├── LICENSE ├── README.md ├── SUMMARY.md ├── book.json ├── g3doc ├── api_docs │ ├── cc │ │ ├── ClassEnv.md │ │ ├── ClassEnvWrapper.md │ │ ├── ClassPartialTensorShape.md │ │ ├── ClassPartialTensorShapeUtils.md │ │ ├── ClassRandomAccessFile.md │ │ ├── ClassSession.md │ │ ├── ClassStatus.md │ │ ├── ClassTensor.md │ │ ├── ClassTensorShape.md │ │ ├── ClassTensorShapeUtils.md │ │ ├── ClassThread.md │ │ ├── ClassWritableFile.md │ │ ├── StructSessionOptions.md │ │ ├── StructState.md │ │ ├── StructTF_Buffer.md │ │ ├── StructTensorShapeDim.md │ │ ├── StructThreadOptions.md │ │ └── index.md │ ├── index.md │ ├── leftnav_files │ └── python │ │ ├── array_ops.md │ │ ├── check_ops.md │ │ ├── client.md │ │ ├── constant_op.md │ │ ├── contrib.copy_graph.md │ │ ├── contrib.distributions.md │ │ ├── contrib.ffmpeg.md │ │ ├── contrib.framework.md │ │ ├── contrib.layers.md │ │ ├── contrib.learn.md │ │ ├── contrib.losses.md │ │ ├── contrib.metrics.md │ │ ├── contrib.util.md │ │ ├── control_flow_ops.md │ │ ├── framework.md │ │ ├── functional_ops.md │ │ ├── functions_and_classes │ │ ├── shard0 │ │ │ ├── tf.ReaderBase.md │ │ │ ├── tf.SparseTensorValue.md │ │ │ ├── tf.TensorArray.md │ │ │ ├── tf.TensorShape.md │ │ │ ├── tf.VariableScope.md │ │ │ ├── tf.add_check_numerics_ops.md │ │ │ ├── tf.add_n.md │ │ │ ├── tf.add_to_collection.md │ │ │ ├── tf.assert_non_positive.md │ │ │ ├── tf.batch_matrix_triangular_solve.md │ │ │ ├── tf.case.md │ │ │ ├── tf.cholesky.md │ │ │ ├── tf.cond.md │ │ │ ├── tf.contrib.distributions.MultivariateNormal.md │ │ │ ├── tf.contrib.framework.get_global_step.md │ │ │ ├── tf.contrib.framework.safe_embedding_lookup_sparse.md │ │ │ ├── tf.contrib.layers.l2_regularizer.md │ │ │ ├── tf.contrib.learn.LinearRegressor.md │ │ │ ├── tf.contrib.learn.extract_pandas_matrix.md │ │ │ ├── tf.contrib.learn.read_batch_record_features.md │ │ │ ├── tf.contrib.losses.get_losses.md │ │ │ ├── tf.contrib.metrics.set_size.md │ │ │ ├── tf.contrib.util.ops_used_by_graph_def.md │ │ │ ├── tf.delete_session_tensor.md │ │ │ ├── tf.depth_to_space.md │ │ │ ├── tf.device.md │ │ │ ├── tf.errors.CancelledError.md │ │ │ ├── tf.errors.DataLossError.md │ │ │ ├── tf.errors.DeadlineExceededError.md │ │ │ ├── tf.fft.md │ │ │ ├── tf.fft2d.md │ │ │ ├── tf.get_seed.md │ │ │ ├── tf.get_session_handle.md │ │ │ ├── tf.global_norm.md │ │ │ ├── tf.ifft3d.md │ │ │ ├── tf.image.adjust_brightness.md │ │ │ ├── tf.image.decode_jpeg.md │ │ │ ├── tf.image.grayscale_to_rgb.md │ │ │ ├── tf.image.random_brightness.md │ │ │ ├── tf.image.rgb_to_grayscale.md │ │ │ ├── tf.is_finite.md │ │ │ ├── tf.is_nan.md │ │ │ ├── tf.is_non_decreasing.md │ │ │ ├── tf.is_numeric_tensor.md │ │ │ ├── tf.mod.md │ │ │ ├── tf.mul.md │ │ │ ├── tf.name_scope.md │ │ │ ├── tf.nn.avg_pool3d.md │ │ │ ├── tf.nn.bidirectional_rnn.md │ │ │ ├── tf.nn.l2_normalize.md │ │ │ ├── tf.nn.rnn.md │ │ │ ├── tf.not_equal.md │ │ │ ├── tf.py_func.md │ │ │ ├── tf.reduce_sum.md │ │ │ ├── tf.reshape.md │ │ │ ├── tf.reverse_sequence.md │ │ │ ├── tf.segment_min.md │ │ │ ├── tf.sparse_tensor_to_dense.md │ │ │ ├── tf.sqrt.md │ │ │ ├── tf.train.ExponentialMovingAverage.md │ │ │ ├── tf.train.add_queue_runner.md │ │ │ ├── tf.train.limit_epochs.md │ │ │ ├── tf.tuple.md │ │ │ └── tf.zeros_initializer.md │ │ ├── shard1 │ │ │ ├── tf.AggregationMethod.md │ │ │ ├── tf.FIFOQueue.md │ │ │ ├── tf.IdentityReader.md │ │ │ ├── tf.NoGradient.md │ │ │ ├── tf.Print.md │ │ │ ├── tf.Tensor.md │ │ │ ├── tf.Variable.from_proto.md │ │ │ ├── tf.accumulate_n.md │ │ │ ├── tf.all_variables.md │ │ │ ├── tf.assert_less_equal.md │ │ │ ├── tf.assert_rank_at_least.md │ │ │ ├── tf.batch_fft.md │ │ │ ├── tf.batch_matrix_inverse.md │ │ │ ├── tf.batch_to_space.md │ │ │ ├── tf.constant_initializer.md │ │ │ ├── tf.contrib.distributions.StudentT.md │ │ │ ├── tf.contrib.framework.get_graph_from_inputs.md │ │ │ ├── tf.contrib.framework.get_local_variables.md │ │ │ ├── tf.contrib.framework.get_variables_by_name.md │ │ │ ├── tf.contrib.layers.summarize_tensors.md │ │ │ ├── tf.contrib.learn.LinearClassifier.md │ │ │ ├── tf.contrib.learn.extract_pandas_data.md │ │ │ ├── tf.contrib.learn.train.md │ │ │ ├── tf.contrib.losses.absolute_difference.md │ │ │ ├── tf.contrib.losses.sum_of_pairwise_squares.md │ │ │ ├── tf.contrib.metrics.streaming_mean_squared_error.md │ │ │ ├── tf.contrib.metrics.streaming_sparse_recall_at_k.md │ │ │ ├── tf.contrib.util.constant_value.md │ │ │ ├── tf.convert_to_tensor.md │ │ │ ├── tf.diag.md │ │ │ ├── tf.erf.md │ │ │ ├── tf.errors.AlreadyExistsError.md │ │ │ ├── tf.get_default_session.md │ │ │ ├── tf.gradients.md │ │ │ ├── tf.greater_equal.md │ │ │ ├── tf.igammac.md │ │ │ ├── tf.image.adjust_hue.md │ │ │ ├── tf.image.central_crop.md │ │ │ ├── tf.image.random_hue.md │ │ │ ├── tf.image.resize_bicubic.md │ │ │ ├── tf.invert_permutation.md │ │ │ ├── tf.is_strictly_increasing.md │ │ │ ├── tf.linspace.md │ │ │ ├── tf.map_fn.md │ │ │ ├── tf.matching_files.md │ │ │ ├── tf.merge_all_summaries.md │ │ │ ├── tf.nn.compute_accidental_hits.md │ │ │ ├── tf.nn.embedding_lookup_sparse.md │ │ │ ├── tf.nn.erosion2d.md │ │ │ ├── tf.nn.moments.md │ │ │ ├── tf.nn.normalize_moments.md │ │ │ ├── tf.nn.sampled_softmax_loss.md │ │ │ ├── tf.parse_single_example.md │ │ │ ├── tf.random_uniform_initializer.md │ │ │ ├── tf.read_file.md │ │ │ ├── tf.reduce_any.md │ │ │ ├── tf.reduce_join.md │ │ │ ├── tf.string_to_hash_bucket_strong.md │ │ │ ├── tf.test.get_temp_dir.md │ │ │ ├── tf.train.AdadeltaOptimizer.md │ │ │ ├── tf.train.ClusterSpec.md │ │ │ ├── tf.train.GradientDescentOptimizer.md │ │ │ ├── tf.train.LooperThread.md │ │ │ ├── tf.train.Server.create_local_server.md │ │ │ ├── tf.train.generate_checkpoint_state_proto.md │ │ │ ├── tf.train.shuffle_batch_join.md │ │ │ ├── tf.zeros.md │ │ │ └── tf.zeros_like.md │ │ ├── shard2 │ │ │ ├── tf.DType.md │ │ │ ├── tf.Graph.md │ │ │ ├── tf.InteractiveSession.md │ │ │ ├── tf.TFRecordReader.md │ │ │ ├── tf.TextLineReader.md │ │ │ ├── tf.WholeFileReader.md │ │ │ ├── tf.assert_non_negative.md │ │ │ ├── tf.batch_matrix_determinant.md │ │ │ ├── tf.cholesky_solve.md │ │ │ ├── tf.constant.md │ │ │ ├── tf.contrib.copy_graph.get_copied_op.md │ │ │ ├── tf.contrib.distributions.Chi2.md │ │ │ ├── tf.contrib.distributions.Uniform.md │ │ │ ├── tf.contrib.framework.arg_scope.md │ │ │ ├── tf.contrib.framework.assert_global_step.md │ │ │ ├── tf.contrib.framework.assert_scalar_int.md │ │ │ ├── tf.contrib.framework.get_unique_variable.md │ │ │ ├── tf.contrib.framework.get_variables_to_restore.md │ │ │ ├── tf.contrib.framework.with_same_shape.md │ │ │ ├── tf.contrib.layers.optimize_loss.md │ │ │ ├── tf.contrib.learn.BaseEstimator.md │ │ │ ├── tf.contrib.learn.ModeKeys.md │ │ │ ├── tf.contrib.learn.TensorFlowDNNRegressor.md │ │ │ ├── tf.contrib.metrics.set_difference.md │ │ │ ├── tf.contrib.metrics.streaming_auc.md │ │ │ ├── tf.contrib.metrics.streaming_mean.md │ │ │ ├── tf.contrib.metrics.streaming_mean_relative_error.md │ │ │ ├── tf.contrib.metrics.streaming_sparse_precision_at_k.md │ │ │ ├── tf.digamma.md │ │ │ ├── tf.edit_distance.md │ │ │ ├── tf.errors.ResourceExhaustedError.md │ │ │ ├── tf.expand_dims.md │ │ │ ├── tf.gather_nd.md │ │ │ ├── tf.get_variable_scope.md │ │ │ ├── tf.identity.md │ │ │ ├── tf.imag.md │ │ │ ├── tf.image.encode_jpeg.md │ │ │ ├── tf.image.per_image_whitening.md │ │ │ ├── tf.image_summary.md │ │ │ ├── tf.initialize_all_variables.md │ │ │ ├── tf.initialize_local_variables.md │ │ │ ├── tf.is_variable_initialized.md │ │ │ ├── tf.matmul.md │ │ │ ├── tf.minimum.md │ │ │ ├── tf.neg.md │ │ │ ├── tf.nn.embedding_lookup.md │ │ │ ├── tf.nn.log_uniform_candidate_sampler.md │ │ │ ├── tf.nn.relu.md │ │ │ ├── tf.random_uniform.md │ │ │ ├── tf.real.md │ │ │ ├── tf.report_uninitialized_variables.md │ │ │ ├── tf.scatter_update.md │ │ │ ├── tf.shape_n.md │ │ │ ├── tf.sin.md │ │ │ ├── tf.sparse_placeholder.md │ │ │ ├── tf.split.md │ │ │ ├── tf.squeeze.md │ │ │ ├── tf.test.assert_equal_graph_def.md │ │ │ ├── tf.test.compute_gradient.md │ │ │ ├── tf.test.is_built_with_cuda.md │ │ │ ├── tf.to_int32.md │ │ │ ├── tf.train.Saver.from_proto.md │ │ │ ├── tf.train.global_step.md │ │ │ ├── tf.train.latest_checkpoint.md │ │ │ ├── tf.train.shuffle_batch.md │ │ │ ├── tf.train.start_queue_runners.md │ │ │ └── tf.uniform_unit_scaling_initializer.md │ │ ├── shard3 │ │ │ ├── tf.RegisterGradient.md │ │ │ ├── tf.SparseTensor.md │ │ │ ├── tf.assert_rank.md │ │ │ ├── tf.assert_type.md │ │ │ ├── tf.batch_ifft3d.md │ │ │ ├── tf.batch_self_adjoint_eig.md │ │ │ ├── tf.ceil.md │ │ │ ├── tf.check_numerics.md │ │ │ ├── tf.complex_abs.md │ │ │ ├── tf.contrib.copy_graph.copy_op_to_graph.md │ │ │ ├── tf.contrib.distributions.ContinuousDistribution.md │ │ │ ├── tf.contrib.distributions.DirichletMultinomial.md │ │ │ ├── tf.contrib.distributions.Exponential.md │ │ │ ├── tf.contrib.distributions.Gamma.md │ │ │ ├── tf.contrib.distributions.normal_conjugates_known_sigma_posterior.md │ │ │ ├── tf.contrib.ffmpeg.encode_audio.md │ │ │ ├── tf.contrib.framework.add_model_variable.md │ │ │ ├── tf.contrib.framework.get_model_variables.md │ │ │ ├── tf.contrib.framework.local_variable.md │ │ │ ├── tf.contrib.layers.summarize_activation.md │ │ │ ├── tf.contrib.layers.variance_scaling_initializer.md │ │ │ ├── tf.contrib.learn.Estimator.md │ │ │ ├── tf.contrib.learn.extract_dask_data.md │ │ │ ├── tf.contrib.losses.add_loss.md │ │ │ ├── tf.contrib.losses.cosine_distance.md │ │ │ ├── tf.contrib.losses.get_regularization_losses.md │ │ │ ├── tf.contrib.metrics.auc_using_histogram.md │ │ │ ├── tf.control_dependencies.md │ │ │ ├── tf.convert_to_tensor_or_indexed_slices.md │ │ │ ├── tf.decode_csv.md │ │ │ ├── tf.decode_raw.md │ │ │ ├── tf.errors.OutOfRangeError.md │ │ │ ├── tf.errors.UnauthenticatedError.md │ │ │ ├── tf.exp.md │ │ │ ├── tf.foldl.md │ │ │ ├── tf.group.md │ │ │ ├── tf.ifft2d.md │ │ │ ├── tf.image.adjust_contrast.md │ │ │ ├── tf.image.adjust_saturation.md │ │ │ ├── tf.image.convert_image_dtype.md │ │ │ ├── tf.image.decode_png.md │ │ │ ├── tf.image.random_contrast.md │ │ │ ├── tf.image.random_saturation.md │ │ │ ├── tf.local_variables.md │ │ │ ├── tf.logical_xor.md │ │ │ ├── tf.multinomial.md │ │ │ ├── tf.nn.dropout.md │ │ │ ├── tf.nn.log_softmax.md │ │ │ ├── tf.nn.max_pool.md │ │ │ ├── tf.nn.rnn_cell.EmbeddingWrapper.md │ │ │ ├── tf.nn.softsign.md │ │ │ ├── tf.pack.md │ │ │ ├── tf.placeholder.md │ │ │ ├── tf.random_shuffle.md │ │ │ ├── tf.reduce_min.md │ │ │ ├── tf.register_tensor_conversion_function.md │ │ │ ├── tf.scalar_summary.md │ │ │ ├── tf.segment_mean.md │ │ │ ├── tf.shape.md │ │ │ ├── tf.sparse_concat.md │ │ │ ├── tf.sparse_softmax.md │ │ │ ├── tf.sparse_split.md │ │ │ ├── tf.stop_gradient.md │ │ │ ├── tf.test.compute_gradient_error.md │ │ │ ├── tf.to_double.md │ │ │ ├── tf.trace.md │ │ │ ├── tf.train.AdagradOptimizer.md │ │ │ ├── tf.train.QueueRunner.md │ │ │ ├── tf.train.Server.md │ │ │ ├── tf.train.batch_join.md │ │ │ ├── tf.train.range_input_producer.md │ │ │ └── tf.unique_with_counts.md │ │ ├── shard4 │ │ │ ├── tf.Assert.md │ │ │ ├── tf.FixedLengthRecordReader.md │ │ │ ├── tf.argmin.md │ │ │ ├── tf.assert_less.md │ │ │ ├── tf.batch_ifft.md │ │ │ ├── tf.batch_matrix_band_part.md │ │ │ ├── tf.batch_matrix_diag_part.md │ │ │ ├── tf.batch_matrix_solve.md │ │ │ ├── tf.clip_by_value.md │ │ │ ├── tf.complex.md │ │ │ ├── tf.contrib.framework.VariableDeviceChooser.md │ │ │ ├── tf.contrib.framework.add_arg_scope.md │ │ │ ├── tf.contrib.learn.DNNClassifier.md │ │ │ ├── tf.contrib.learn.RunConfig.md │ │ │ ├── tf.contrib.learn.TensorFlowEstimator.md │ │ │ ├── tf.contrib.learn.TensorFlowLinearClassifier.md │ │ │ ├── tf.contrib.learn.TensorFlowRNNRegressor.md │ │ │ ├── tf.contrib.learn.evaluate.md │ │ │ ├── tf.contrib.learn.read_batch_features.md │ │ │ ├── tf.contrib.learn.run_feeds.md │ │ │ ├── tf.contrib.metrics.set_union.md │ │ │ ├── tf.contrib.metrics.streaming_recall_at_k.md │ │ │ ├── tf.contrib.util.stripped_op_list_for_graph.md │ │ │ ├── tf.errors.FailedPreconditionError.md │ │ │ ├── tf.extract_image_patches.md │ │ │ ├── tf.floor.md │ │ │ ├── tf.greater.md │ │ │ ├── tf.histogram_fixed_width.md │ │ │ ├── tf.image.hsv_to_rgb.md │ │ │ ├── tf.initialize_variables.md │ │ │ ├── tf.log.md │ │ │ ├── tf.nn.conv2d_transpose.md │ │ │ ├── tf.nn.depthwise_conv2d.md │ │ │ ├── tf.nn.depthwise_conv2d_native.md │ │ │ ├── tf.nn.dilation2d.md │ │ │ ├── tf.nn.l2_loss.md │ │ │ ├── tf.nn.max_pool3d.md │ │ │ ├── tf.nn.nce_loss.md │ │ │ ├── tf.nn.rnn_cell.OutputProjectionWrapper.md │ │ │ ├── tf.nn.softplus.md │ │ │ ├── tf.nn.sparse_softmax_cross_entropy_with_logits.md │ │ │ ├── tf.placeholder_with_default.md │ │ │ ├── tf.reduce_all.md │ │ │ ├── tf.reduce_mean.md │ │ │ ├── tf.segment_max.md │ │ │ ├── tf.select.md │ │ │ ├── tf.sparse_add.md │ │ │ ├── tf.sparse_to_indicator.md │ │ │ ├── tf.string_to_hash_bucket_fast.md │ │ │ ├── tf.sub.md │ │ │ ├── tf.tile.md │ │ │ ├── tf.train.match_filenames_once.md │ │ │ ├── tf.train.update_checkpoint_state.md │ │ │ ├── tf.train.write_graph.md │ │ │ ├── tf.unpack.md │ │ │ ├── tf.unsorted_segment_sum.md │ │ │ └── tf.while_loop.md │ │ ├── shard5 │ │ │ ├── tf.OpError.md │ │ │ ├── tf.RandomShuffleQueue.md │ │ │ ├── tf.add.md │ │ │ ├── tf.asin.md │ │ │ ├── tf.assert_integer.md │ │ │ ├── tf.batch_cholesky_solve.md │ │ │ ├── tf.batch_matmul.md │ │ │ ├── tf.boolean_mask.md │ │ │ ├── tf.bytes.md │ │ │ ├── tf.cast.md │ │ │ ├── tf.clip_by_global_norm.md │ │ │ ├── tf.clip_by_norm.md │ │ │ ├── tf.contrib.framework.arg_scoped_arguments.md │ │ │ ├── tf.contrib.framework.get_variables.md │ │ │ ├── tf.contrib.framework.variable.md │ │ │ ├── tf.contrib.layers.convolution2d.md │ │ │ ├── tf.contrib.layers.fully_connected.md │ │ │ ├── tf.contrib.layers.sum_regularizer.md │ │ │ ├── tf.contrib.layers.summarize_collection.md │ │ │ ├── tf.contrib.learn.TensorFlowDNNClassifier.md │ │ │ ├── tf.contrib.learn.run_n.md │ │ │ ├── tf.contrib.losses.sigmoid_cross_entropy.md │ │ │ ├── tf.contrib.metrics.accuracy.md │ │ │ ├── tf.contrib.metrics.streaming_percentage_less.md │ │ │ ├── tf.cos.md │ │ │ ├── tf.diag_part.md │ │ │ ├── tf.div.md │ │ │ ├── tf.errors.PermissionDeniedError.md │ │ │ ├── tf.errors.UnavailableError.md │ │ │ ├── tf.floordiv.md │ │ │ ├── tf.histogram_summary.md │ │ │ ├── tf.image.flip_left_right.md │ │ │ ├── tf.image.flip_up_down.md │ │ │ ├── tf.image.pad_to_bounding_box.md │ │ │ ├── tf.image.resize_image_with_crop_or_pad.md │ │ │ ├── tf.image.resize_nearest_neighbor.md │ │ │ ├── tf.image.sample_distorted_bounding_box.md │ │ │ ├── tf.inv.md │ │ │ ├── tf.listdiff.md │ │ │ ├── tf.load_file_system_library.md │ │ │ ├── tf.logical_and.md │ │ │ ├── tf.logical_not.md │ │ │ ├── tf.make_template.md │ │ │ ├── tf.merge_summary.md │ │ │ ├── tf.nn.conv3d.md │ │ │ ├── tf.nn.rnn_cell.BasicRNNCell.md │ │ │ ├── tf.nn.rnn_cell.DropoutWrapper.md │ │ │ ├── tf.nn.rnn_cell.GRUCell.md │ │ │ ├── tf.nn.rnn_cell.InputProjectionWrapper.md │ │ │ ├── tf.nn.rnn_cell.LSTMCell.md │ │ │ ├── tf.nn.sigmoid_cross_entropy_with_logits.md │ │ │ ├── tf.nn.state_saving_rnn.md │ │ │ ├── tf.nn.top_k.md │ │ │ ├── tf.no_op.md │ │ │ ├── tf.range.md │ │ │ ├── tf.sign.md │ │ │ ├── tf.sparse_segment_sqrt_n_grad.md │ │ │ ├── tf.tan.md │ │ │ ├── tf.test.main.md │ │ │ ├── tf.to_int64.md │ │ │ ├── tf.train.FtrlOptimizer.md │ │ │ ├── tf.train.LooperThread.loop.md │ │ │ ├── tf.train.Optimizer.md │ │ │ ├── tf.train.Saver.md │ │ │ ├── tf.train.SessionManager.md │ │ │ ├── tf.train.replica_device_setter.md │ │ │ ├── tf.transpose.md │ │ │ ├── tf.truncated_normal_initializer.md │ │ │ ├── tf.variable_op_scope.md │ │ │ └── tf.verify_tensor_all_finite.md │ │ ├── shard6 │ │ │ ├── tf.Dimension.md │ │ │ ├── tf.FixedLenSequenceFeature.md │ │ │ ├── tf.PaddingFIFOQueue.md │ │ │ ├── tf.QueueBase.md │ │ │ ├── tf.abs.md │ │ │ ├── tf.assert_positive.md │ │ │ ├── tf.batch_fft3d.md │ │ │ ├── tf.batch_ifft2d.md │ │ │ ├── tf.bitcast.md │ │ │ ├── tf.concat.md │ │ │ ├── tf.conj.md │ │ │ ├── tf.contrib.copy_graph.copy_variable_to_graph.md │ │ │ ├── tf.contrib.framework.create_global_step.md │ │ │ ├── tf.contrib.framework.reduce_sum_n.md │ │ │ ├── tf.contrib.layers.summarize_activations.md │ │ │ ├── tf.contrib.learn.extract_dask_labels.md │ │ │ ├── tf.contrib.learn.read_batch_examples.md │ │ │ ├── tf.contrib.losses.log_loss.md │ │ │ ├── tf.contrib.losses.sum_of_squares.md │ │ │ ├── tf.contrib.metrics.set_intersection.md │ │ │ ├── tf.contrib.metrics.streaming_recall.md │ │ │ ├── tf.contrib.metrics.streaming_root_mean_squared_error.md │ │ │ ├── tf.contrib.util.make_tensor_proto.md │ │ │ ├── tf.decode_json_example.md │ │ │ ├── tf.dynamic_partition.md │ │ │ ├── tf.erfc.md │ │ │ ├── tf.errors.AbortedError.md │ │ │ ├── tf.errors.InternalError.md │ │ │ ├── tf.errors.NotFoundError.md │ │ │ ├── tf.errors.UnimplementedError.md │ │ │ ├── tf.igamma.md │ │ │ ├── tf.image.extract_glimpse.md │ │ │ ├── tf.image.rgb_to_hsv.md │ │ │ ├── tf.import_graph_def.md │ │ │ ├── tf.load_op_library.md │ │ │ ├── tf.maximum.md │ │ │ ├── tf.moving_average_variables.md │ │ │ ├── tf.nn.elu.md │ │ │ ├── tf.nn.separable_conv2d.md │ │ │ ├── tf.nn.softmax.md │ │ │ ├── tf.one_hot.md │ │ │ ├── tf.op_scope.md │ │ │ ├── tf.parse_example.md │ │ │ ├── tf.pow.md │ │ │ ├── tf.python_io.tf_record_iterator.md │ │ │ ├── tf.random_crop.md │ │ │ ├── tf.random_normal_initializer.md │ │ │ ├── tf.rank.md │ │ │ ├── tf.self_adjoint_eig.md │ │ │ ├── tf.sigmoid.md │ │ │ ├── tf.slice.md │ │ │ ├── tf.space_to_depth.md │ │ │ ├── tf.sparse_reset_shape.md │ │ │ ├── tf.sparse_segment_mean.md │ │ │ ├── tf.sparse_segment_sum.md │ │ │ ├── tf.to_float.md │ │ │ ├── tf.train.AdamOptimizer.md │ │ │ ├── tf.train.Coordinator.md │ │ │ ├── tf.train.QueueRunner.from_proto.md │ │ │ ├── tf.train.RMSPropOptimizer.md │ │ │ ├── tf.train.Supervisor.md │ │ │ ├── tf.train.exponential_decay.md │ │ │ ├── tf.train.slice_input_producer.md │ │ │ ├── tf.trainable_variables.md │ │ │ └── tf.truncated_normal.md │ │ ├── shard7 │ │ │ ├── tf.DeviceSpec.from_string.md │ │ │ ├── tf.FixedLenFeature.md │ │ │ ├── tf.Operation.md │ │ │ ├── tf.QueueBase.from_list.md │ │ │ ├── tf.as_dtype.md │ │ │ ├── tf.assert_equal.md │ │ │ ├── tf.assert_variables_initialized.md │ │ │ ├── tf.batch_cholesky.md │ │ │ ├── tf.contrib.distributions.BaseDistribution.md │ │ │ ├── tf.contrib.distributions.DiscreteDistribution.md │ │ │ ├── tf.contrib.distributions.Normal.md │ │ │ ├── tf.contrib.framework.assert_or_get_global_step.md │ │ │ ├── tf.contrib.framework.get_variables_by_suffix.md │ │ │ ├── tf.contrib.framework.has_arg_scope.md │ │ │ ├── tf.contrib.framework.with_shape.md │ │ │ ├── tf.contrib.learn.TensorFlowClassifier.md │ │ │ ├── tf.contrib.losses.softmax_cross_entropy.md │ │ │ ├── tf.contrib.metrics.streaming_mean_absolute_error.md │ │ │ ├── tf.contrib.metrics.streaming_precision.md │ │ │ ├── tf.contrib.util.make_ndarray.md │ │ │ ├── tf.count_up_to.md │ │ │ ├── tf.dynamic_stitch.md │ │ │ ├── tf.fft3d.md │ │ │ ├── tf.get_collection.md │ │ │ ├── tf.get_collection_ref.md │ │ │ ├── tf.get_session_tensor.md │ │ │ ├── tf.ifft.md │ │ │ ├── tf.image.random_flip_left_right.md │ │ │ ├── tf.image.resize_area.md │ │ │ ├── tf.less.md │ │ │ ├── tf.lgamma.md │ │ │ ├── tf.logical_or.md │ │ │ ├── tf.matrix_solve.md │ │ │ ├── tf.matrix_solve_ls.md │ │ │ ├── tf.nn.atrous_conv2d.md │ │ │ ├── tf.nn.avg_pool.md │ │ │ ├── tf.nn.fixed_unigram_candidate_sampler.md │ │ │ ├── tf.nn.in_top_k.md │ │ │ ├── tf.nn.local_response_normalization.md │ │ │ ├── tf.nn.rnn_cell.BasicLSTMCell.md │ │ │ ├── tf.nn.softmax_cross_entropy_with_logits.md │ │ │ ├── tf.ones_like.md │ │ │ ├── tf.polygamma.md │ │ │ ├── tf.python_io.TFRecordWriter.md │ │ │ ├── tf.random_normal.md │ │ │ ├── tf.reduce_max.md │ │ │ ├── tf.scatter_sub.md │ │ │ ├── tf.segment_prod.md │ │ │ ├── tf.segment_sum.md │ │ │ ├── tf.sparse_reduce_sum.md │ │ │ ├── tf.sparse_tensor_dense_matmul.md │ │ │ ├── tf.square.md │ │ │ ├── tf.string_to_hash_bucket.md │ │ │ ├── tf.train.MomentumOptimizer.md │ │ │ ├── tf.train.SummaryWriter.md │ │ │ ├── tf.train.export_meta_graph.md │ │ │ ├── tf.train.get_checkpoint_state.md │ │ │ ├── tf.truediv.md │ │ │ ├── tf.unique.md │ │ │ ├── tf.variable_scope.md │ │ │ └── tf.where.md │ │ ├── shard8 │ │ │ ├── tf.GraphKeys.md │ │ │ ├── tf.IndexedSlices.md │ │ │ ├── tf.RegisterShape.md │ │ │ ├── tf.Session.md │ │ │ ├── tf.VarLenFeature.md │ │ │ ├── tf.Variable.md │ │ │ ├── tf.acos.md │ │ │ ├── tf.argmax.md │ │ │ ├── tf.assert_negative.md │ │ │ ├── tf.assert_proper_iterable.md │ │ │ ├── tf.atan.md │ │ │ ├── tf.batch_matrix_diag.md │ │ │ ├── tf.batch_matrix_solve_ls.md │ │ │ ├── tf.contrib.distributions.normal_congugates_known_sigma_predictive.md │ │ │ ├── tf.contrib.framework.assert_same_float_dtype.md │ │ │ ├── tf.contrib.layers.apply_regularization.md │ │ │ ├── tf.contrib.layers.summarize_tensor.md │ │ │ ├── tf.contrib.layers.xavier_initializer_conv2d.md │ │ │ ├── tf.contrib.learn.NanLossDuringTrainingError.md │ │ │ ├── tf.contrib.learn.TensorFlowRegressor.md │ │ │ ├── tf.contrib.learn.extract_pandas_labels.md │ │ │ ├── tf.contrib.metrics.confusion_matrix.md │ │ │ ├── tf.contrib.metrics.streaming_accuracy.md │ │ │ ├── tf.contrib.metrics.streaming_mean_cosine_distance.md │ │ │ ├── tf.cross.md │ │ │ ├── tf.equal.md │ │ │ ├── tf.image.crop_to_bounding_box.md │ │ │ ├── tf.image.draw_bounding_boxes.md │ │ │ ├── tf.image.resize_bilinear.md │ │ │ ├── tf.image.resize_images.md │ │ │ ├── tf.image.transpose_image.md │ │ │ ├── tf.is_inf.md │ │ │ ├── tf.lbeta.md │ │ │ ├── tf.less_equal.md │ │ │ ├── tf.matrix_inverse.md │ │ │ ├── tf.nn.batch_normalization.md │ │ │ ├── tf.nn.conv2d.md │ │ │ ├── tf.nn.dynamic_rnn.md │ │ │ ├── tf.nn.learned_unigram_candidate_sampler.md │ │ │ ├── tf.nn.relu6.md │ │ │ ├── tf.nn.rnn_cell.LSTMStateTuple.md │ │ │ ├── tf.nn.rnn_cell.MultiRNNCell.md │ │ │ ├── tf.nn.sufficient_statistics.md │ │ │ ├── tf.nn.weighted_cross_entropy_with_logits.md │ │ │ ├── tf.no_regularizer.md │ │ │ ├── tf.ones_initializer.md │ │ │ ├── tf.reduce_prod.md │ │ │ ├── tf.reset_default_graph.md │ │ │ ├── tf.reverse.md │ │ │ ├── tf.round.md │ │ │ ├── tf.rsqrt.md │ │ │ ├── tf.scatter_add.md │ │ │ ├── tf.set_random_seed.md │ │ │ ├── tf.sparse_fill_empty_rows.md │ │ │ ├── tf.sparse_reorder.md │ │ │ ├── tf.sparse_retain.md │ │ │ ├── tf.sparse_segment_sqrt_n.md │ │ │ ├── tf.sparse_to_dense.md │ │ │ ├── tf.train.batch.md │ │ │ ├── tf.train.import_meta_graph.md │ │ │ ├── tf.variable_axis_size_partitioner.md │ │ │ └── tf.zeta.md │ │ └── shard9 │ │ │ ├── tf.DeviceSpec.md │ │ │ ├── tf.audio_summary.md │ │ │ ├── tf.batch_fft2d.md │ │ │ ├── tf.clip_by_average_norm.md │ │ │ ├── tf.contrib.ffmpeg.decode_audio.md │ │ │ ├── tf.contrib.framework.convert_to_tensor_or_sparse_tensor.md │ │ │ ├── tf.contrib.framework.get_or_create_global_step.md │ │ │ ├── tf.contrib.framework.model_variable.md │ │ │ ├── tf.contrib.layers.l1_regularizer.md │ │ │ ├── tf.contrib.layers.xavier_initializer.md │ │ │ ├── tf.contrib.learn.DNNRegressor.md │ │ │ ├── tf.contrib.learn.TensorFlowLinearRegressor.md │ │ │ ├── tf.contrib.learn.TensorFlowRNNClassifier.md │ │ │ ├── tf.contrib.learn.infer.md │ │ │ ├── tf.contrib.losses.get_total_loss.md │ │ │ ├── tf.errors.InvalidArgumentError.md │ │ │ ├── tf.errors.UnknownError.md │ │ │ ├── tf.fill.md │ │ │ ├── tf.foldr.md │ │ │ ├── tf.gather.md │ │ │ ├── tf.get_default_graph.md │ │ │ ├── tf.get_variable.md │ │ │ ├── tf.image.encode_png.md │ │ │ ├── tf.image.random_flip_up_down.md │ │ │ ├── tf.matrix_determinant.md │ │ │ ├── tf.matrix_triangular_solve.md │ │ │ ├── tf.nn.bias_add.md │ │ │ ├── tf.nn.max_pool_with_argmax.md │ │ │ ├── tf.nn.rnn_cell.RNNCell.md │ │ │ ├── tf.nn.uniform_candidate_sampler.md │ │ │ ├── tf.nn.zero_fraction.md │ │ │ ├── tf.ones.md │ │ │ ├── tf.pad.md │ │ │ ├── tf.saturate_cast.md │ │ │ ├── tf.scalar_mul.md │ │ │ ├── tf.scan.md │ │ │ ├── tf.size.md │ │ │ ├── tf.space_to_batch.md │ │ │ ├── tf.sparse_mask.md │ │ │ ├── tf.sparse_merge.md │ │ │ ├── tf.squared_difference.md │ │ │ ├── tf.string_to_number.md │ │ │ ├── tf.tanh.md │ │ │ ├── tf.to_bfloat16.md │ │ │ ├── tf.train.input_producer.md │ │ │ ├── tf.train.string_input_producer.md │ │ │ └── tf.train.summary_iterator.md │ │ ├── histogram_ops.md │ │ ├── image.md │ │ ├── index.md │ │ ├── io_ops.md │ │ ├── math_ops.md │ │ ├── nn.md │ │ ├── python_io.md │ │ ├── rnn_cell.md │ │ ├── script_ops.md │ │ ├── session_ops.md │ │ ├── sparse_ops.md │ │ ├── state_ops.md │ │ ├── string_ops.md │ │ ├── tensor_array_ops.md │ │ ├── test.md │ │ └── train.md ├── contrib │ └── learn │ │ ├── get_started │ │ └── index.md │ │ └── index.md ├── extras │ └── README.txt ├── get_started │ ├── basic_usage.md │ ├── index.md │ ├── leftnav_files │ └── os_setup.md ├── how_tos │ ├── __init__.py │ ├── adding_an_op │ │ ├── BUILD │ │ ├── __init__.py │ │ ├── attr_examples.cc │ │ ├── cuda_op.py │ │ ├── cuda_op_kernel.cc │ │ ├── cuda_op_kernel.cu.cc │ │ ├── cuda_op_test.py │ │ ├── fact_test.py │ │ ├── index.md │ │ ├── zero_out_1_test.py │ │ ├── zero_out_2_test.py │ │ ├── zero_out_3_test.py │ │ ├── zero_out_grad_2.py │ │ ├── zero_out_op_1.py │ │ ├── zero_out_op_2.py │ │ ├── zero_out_op_3.py │ │ ├── zero_out_op_kernel_1.cc │ │ ├── zero_out_op_kernel_2.cc │ │ └── zero_out_op_kernel_3.cc │ ├── distributed │ │ └── index.md │ ├── documentation │ │ └── index.md │ ├── graph_viz │ │ └── index.md │ ├── hadoop │ │ └── index.md │ ├── image_retraining │ │ └── index.md │ ├── index.md │ ├── leftnav_files │ ├── meta_graph │ │ └── index.md │ ├── new_data_formats │ │ └── index.md │ ├── quantization │ │ └── index.md │ ├── reading_data │ │ └── index.md │ ├── style_guide.md │ ├── summaries_and_tensorboard │ │ └── index.md │ ├── threading_and_queues │ │ └── index.md │ ├── tool_developers │ │ └── index.md │ ├── using_gpu │ │ └── index.md │ ├── variable_scope │ │ └── index.md │ └── variables │ │ └── index.md ├── images │ ├── AlexClassification.png │ ├── AnimatedFileQueues.gif │ ├── DynamicPartition.png │ ├── DynamicStitch.png │ ├── Gather.png │ ├── IncremeterFifoQueue.gif │ ├── MNIST-Matrix.png │ ├── MNIST.png │ ├── Parallelism.png │ ├── audio-image-text.png │ ├── baseball_network.png │ ├── blue_pill.png │ ├── cifar_activations.png │ ├── cifar_graph.png │ ├── cifar_image_summary.png │ ├── cifar_loss.png │ ├── cifar_lr_decay.png │ ├── cifar_samples.png │ ├── cifar_sparsity.png │ ├── cifar_var_histograms.png │ ├── colorby_compute_time.png │ ├── colorby_device.png │ ├── colorby_structure.png │ ├── constant.png │ ├── control_edge.png │ ├── conv_1.png │ ├── cropped_panda.jpg │ ├── dataflow_edge.png │ ├── getting_started.dot │ ├── getting_started.png │ ├── grace_hopper.jpg │ ├── graph_vis_animation.gif │ ├── horizontal_stack.png │ ├── infocard.png │ ├── infocard_op.png │ ├── iris_three_species.jpg │ ├── linear-relationships.png │ ├── mandelbrot_output.jpg │ ├── mnist-train-xs.png │ ├── mnist-train-ys.png │ ├── mnist1.png │ ├── mnist10.png │ ├── mnist2.png │ ├── mnist3.png │ ├── mnist4.png │ ├── mnist5.png │ ├── mnist6.png │ ├── mnist7.png │ ├── mnist8.png │ ├── mnist9.png │ ├── mnist_digits.png │ ├── mnist_subgraph.png │ ├── mnist_tensorboard.png │ ├── namespace_node.png │ ├── nce-nplm.png │ ├── op_node.png │ ├── pde_output_1.jpg │ ├── pde_output_2.jpg │ ├── pool1_collapsed.png │ ├── pool1_expanded.png │ ├── re.png │ ├── re1.png │ ├── red_pill.png │ ├── reference_edge.png │ ├── results.png │ ├── run_metadata_graph.png │ ├── run_metadata_infocard.png │ ├── rw4.png │ ├── rw5.png │ ├── rw6.png │ ├── rw7.png │ ├── save.png │ ├── scatterplot.png │ ├── series.png │ ├── series_expanded.png │ ├── softmax-nplm.png │ ├── softmax-regression-scalarequation.png │ ├── softmax-regression-scalargraph.png │ ├── softmax-regression-vectorequation.png │ ├── softmax-weights.png │ ├── summary.png │ ├── tensor_shapes.png │ ├── tensors_flowing.gif │ ├── tf_logo.png │ ├── tf_logo_transp.png │ ├── theta.png │ ├── tsne.png │ ├── vertical_stack.png │ ├── vr1.png │ ├── vr2.png │ ├── vr3.png │ ├── vr4.png │ ├── vr5.png │ ├── wide_n_deep.svg │ └── word2vec2.png ├── index.md ├── resources │ ├── bib.md │ ├── dims_types.md │ ├── faq.md │ ├── glossary.md │ ├── index.md │ ├── leftnav_files │ ├── roadmap.md │ ├── uses.md │ └── versions.md └── tutorials │ ├── BUILD │ ├── __init__.py │ ├── deep_cnn │ ├── cifar_tensorboard.html │ └── index.md │ ├── image_recognition │ └── index.md │ ├── index.md │ ├── leftnav_files │ ├── linear │ └── overview.md │ ├── mandelbrot │ └── index.md │ ├── mnist │ ├── beginners │ │ └── index.md │ ├── download │ │ └── index.md │ ├── pros │ │ └── index.md │ └── tf │ │ └── index.md │ ├── monitors │ └── index.md │ ├── pdes │ └── index.md │ ├── recurrent │ └── index.md │ ├── seq2seq │ └── index.md │ ├── syntaxnet │ └── index.md │ ├── tflearn │ └── index.md │ ├── tfserve │ └── index.md │ ├── wide │ └── index.md │ ├── wide_and_deep │ └── index.md │ └── word2vec │ └── index.md └── progress.md /book.json: -------------------------------------------------------------------------------- 1 | { 2 | "plugins": ["mathjax"] 3 | } 4 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/ClassPartialTensorShapeUtils.md: -------------------------------------------------------------------------------- 1 | # `class tensorflow::PartialTensorShapeUtils` 2 | 3 | Static helper routines for ` PartialTensorShape `. Includes a few common predicates on a partially known tensor shape. 4 | 5 | 6 | 7 | ###Member Details 8 | 9 | #### `string tensorflow::PartialTensorShapeUtils::PartialShapeListString(const gtl::ArraySlice< PartialTensorShape > &shapes)` {#string_tensorflow_PartialTensorShapeUtils_PartialShapeListString} 10 | 11 | 12 | 13 | 14 | 15 | #### `bool tensorflow::PartialTensorShapeUtils::AreCompatible(const gtl::ArraySlice< PartialTensorShape > &shapes0, const gtl::ArraySlice< PartialTensorShape > &shapes1)` {#bool_tensorflow_PartialTensorShapeUtils_AreCompatible} 16 | 17 | 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/ClassThread.md: -------------------------------------------------------------------------------- 1 | # `class tensorflow::Thread` 2 | 3 | 4 | 5 | 6 | 7 | ###Member Details 8 | 9 | #### `tensorflow::Thread::Thread()` {#tensorflow_Thread_Thread} 10 | 11 | 12 | 13 | 14 | 15 | #### `tensorflow::Thread::~Thread()` {#tensorflow_Thread_Thread} 16 | 17 | Blocks until the thread of control stops running. 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/StructState.md: -------------------------------------------------------------------------------- 1 | # `struct tensorflow::Status::State` 2 | 3 | 4 | 5 | 6 | 7 | ###Member Details 8 | 9 | #### `tensorflow::error::Code tensorflow::Status::State::code` {#tensorflow_error_Code_tensorflow_Status_State_code} 10 | 11 | 12 | 13 | 14 | 15 | #### `string tensorflow::Status::State::msg` {#string_tensorflow_Status_State_msg} 16 | 17 | 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/StructTF_Buffer.md: -------------------------------------------------------------------------------- 1 | # `struct TF_Buffer` 2 | 3 | 4 | 5 | 6 | 7 | ###Member Details 8 | 9 | #### `const void* TF_Buffer::data` {#const_void_TF_Buffer_data} 10 | 11 | 12 | 13 | 14 | 15 | #### `size_t TF_Buffer::length` {#size_t_TF_Buffer_length} 16 | 17 | 18 | 19 | 20 | 21 | #### `void(* TF_Buffer::data_deallocator) (void *data, size_t length))(void *data, size_t length)` {#void_TF_Buffer_data_deallocator_void_data_size_t_length_} 22 | 23 | 24 | 25 | 26 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/StructTensorShapeDim.md: -------------------------------------------------------------------------------- 1 | # `struct tensorflow::TensorShapeDim` 2 | 3 | 4 | 5 | 6 | 7 | ###Member Details 8 | 9 | #### `int64 tensorflow::TensorShapeDim::size` {#int64_tensorflow_TensorShapeDim_size} 10 | 11 | 12 | 13 | 14 | 15 | #### `tensorflow::TensorShapeDim::TensorShapeDim(int64 s)` {#tensorflow_TensorShapeDim_TensorShapeDim} 16 | 17 | 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/cc/StructThreadOptions.md: -------------------------------------------------------------------------------- 1 | # `struct tensorflow::ThreadOptions` 2 | 3 | Options to configure a Thread . 4 | 5 | Note that the options are all hints, and the underlying implementation may choose to ignore it. 6 | 7 | ###Member Details 8 | 9 | #### `size_t tensorflow::ThreadOptions::stack_size` {#size_t_tensorflow_ThreadOptions_stack_size} 10 | 11 | Thread stack size to use (in bytes). 12 | 13 | 14 | 15 | #### `size_t tensorflow::ThreadOptions::guard_size` {#size_t_tensorflow_ThreadOptions_guard_size} 16 | 17 | Guard area size to use near thread stacks to use (in bytes) 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/index.md: -------------------------------------------------------------------------------- 1 | # API 개요 2 | 3 | 텐서플로우(TensorFlow)는 그래프를 설계하고 실행하기 위해 여러 개발언어에서 사용 가능한 4 | API들을 가지고 있습니다. 파이썬 API는 현재 가장 완벽하며 사용하기 쉽고, 안드로이드와 5 | 같은 모바일 기기에 배포하는 것을 지원합니다. 하지만 C++ API가 그래프를 실행하는데 6 | 있어 더 성능이 뛰어날 수 있습니다. 7 | 8 | 시간이 지나게 되면, 우리는 텐서플로우 커뮤니티 유저들이 Go, Java, Javascript, Lua, R 9 | 같은 언어 에서도 프론트엔드 개발할 수 있기를 바랍니다. [SWIG](http://swig.org)을 사용한다면, 10 | 비교적 쉽게 텐서플로우 인터페이스를 가장 좋아하는 언어로 개발할 수 있습니다. 11 | 12 | 노트: 대부분의 실제 사용사례는 Machanics 탭에 수록되어 있으며, 13 | 특정 언어 API에 제한되어 있지 않은 추가적인 문서는 Resources 탭에서 확인하실 수 있습니다. 14 | 15 | * [Python API](python/index.md) 16 | * [C++ API](cc/index.md) 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.SparseTensorValue.md: -------------------------------------------------------------------------------- 1 | SparseTensorValue(indices, values, shape) 2 | - - - 3 | 4 | #### `tf.SparseTensorValue.indices` {#SparseTensorValue.indices} 5 | 6 | Alias for field number 0 7 | 8 | 9 | - - - 10 | 11 | #### `tf.SparseTensorValue.shape` {#SparseTensorValue.shape} 12 | 13 | Alias for field number 2 14 | 15 | 16 | - - - 17 | 18 | #### `tf.SparseTensorValue.values` {#SparseTensorValue.values} 19 | 20 | Alias for field number 1 21 | 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.add_check_numerics_ops.md: -------------------------------------------------------------------------------- 1 | ### `tf.add_check_numerics_ops()` {#add_check_numerics_ops} 2 | 3 | Connect a `check_numerics` to every floating point tensor. 4 | 5 | `check_numerics` operations themselves are added for each `float` or `double` 6 | tensor in the graph. For all ops in the graph, the `check_numerics` op for 7 | all of its (`float` or `double`) inputs is guaranteed to run before the 8 | `check_numerics` op on any of its outputs. 9 | 10 | ##### Returns: 11 | 12 | A `group` op depending on all `check_numerics` ops added. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.add_n.md: -------------------------------------------------------------------------------- 1 | ### `tf.add_n(inputs, name=None)` {#add_n} 2 | 3 | Add all input tensors element wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `inputs`: A list of at least 1 `Tensor` objects of the same type in: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 9 | Must all be the same size and shape. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `inputs`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.add_to_collection.md: -------------------------------------------------------------------------------- 1 | ### `tf.add_to_collection(name, value)` {#add_to_collection} 2 | 3 | Wrapper for `Graph.add_to_collection()` using the default graph. 4 | 5 | See [`Graph.add_to_collection()`](../../api_docs/python/framework.md#Graph.add_to_collection) 6 | for more details. 7 | 8 | ##### Args: 9 | 10 | 11 | * `name`: The key for the collection. For example, the `GraphKeys` class 12 | contains many standard names for collections. 13 | * `value`: The value to add to the collection. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.cholesky.md: -------------------------------------------------------------------------------- 1 | ### `tf.cholesky(input, name=None)` {#cholesky} 2 | 3 | Calculates the Cholesky decomposition of a square matrix. 4 | 5 | The input has to be symmetric and positive definite. Only the lower-triangular 6 | part of the input will be used for this operation. The upper-triangular part 7 | will not be read. 8 | 9 | The result is the lower-triangular matrix of the Cholesky decomposition of the 10 | input, `L`, so that `input = L L^*`. 11 | 12 | ##### Args: 13 | 14 | 15 | * `input`: A `Tensor`. Must be one of the following types: `float64`, `float32`. 16 | Shape is `[M, M]`. 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A `Tensor`. Has the same type as `input`. Shape is `[M, M]`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.framework.get_global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_global_step(graph=None)` {#get_global_step} 2 | 3 | Get the global step tensor. 4 | 5 | The global step tensor must be an integer variable. We first try to find it 6 | in the collection `GLOBAL_STEP`, or by name `global_step:0`. 7 | 8 | ##### Args: 9 | 10 | 11 | * `graph`: The graph to find the global step in. If missing, use default graph. 12 | 13 | ##### Returns: 14 | 15 | The global step variable, or `None` if none was found. 16 | 17 | ##### Raises: 18 | 19 | 20 | * `TypeError`: If the global step tensor has a non-integer type, or if it is not 21 | a `Variable`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.layers.l2_regularizer.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.l2_regularizer(scale)` {#l2_regularizer} 2 | 3 | Returns a function that can be used to apply L2 regularization to weights. 4 | 5 | Small values of L2 can help prevent overfitting the training data. 6 | 7 | ##### Args: 8 | 9 | 10 | * `scale`: A scalar multiplier `Tensor`. 0.0 disables the regularizer. 11 | 12 | ##### Returns: 13 | 14 | A function with signature `l2(weights, name=None)` that applies L2 15 | regularization. 16 | 17 | ##### Raises: 18 | 19 | 20 | * `ValueError`: If scale is outside of the range [0.0, 1.0] or if scale is not a 21 | float. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.learn.extract_pandas_matrix.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.extract_pandas_matrix(data)` {#extract_pandas_matrix} 2 | 3 | Extracts numpy matrix from pandas DataFrame. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.losses.get_losses.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.losses.get_losses(scope=None)` {#get_losses} 2 | 3 | Gets the list of loss variables. 4 | 5 | ##### Args: 6 | 7 | 8 | * `scope`: an optional scope for filtering the losses to return. 9 | 10 | ##### Returns: 11 | 12 | a list of loss variables. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.metrics.set_size.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.metrics.set_size(a, validate_indices=True)` {#set_size} 2 | 3 | Compute number of unique elements along last dimension of `a`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `a`: `SparseTensor`, with indices sorted in row-major order. 9 | * `validate_indices`: Whether to validate the order and range of sparse indices 10 | in `a`. 11 | 12 | ##### Returns: 13 | 14 | For `a` ranked `n`, this is a `Tensor` with rank `n-1`, and the same 1st 15 | `n-1` dimensions as `a`. Each value is the number of unique elements in 16 | the corresponding `[0...n-1]` dimension of `a`. 17 | 18 | ##### Raises: 19 | 20 | 21 | * `TypeError`: If `a` is an invalid types. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.contrib.util.ops_used_by_graph_def.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.util.ops_used_by_graph_def(graph_def)` {#ops_used_by_graph_def} 2 | 3 | Collect the list of ops used by a graph. 4 | 5 | Does not validate that the ops are all registered. 6 | 7 | ##### Args: 8 | 9 | 10 | * `graph_def`: A `GraphDef` proto, as from `graph.as_graph_def()`. 11 | 12 | ##### Returns: 13 | 14 | A list of strings, each naming an op used by the graph. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.delete_session_tensor.md: -------------------------------------------------------------------------------- 1 | ### `tf.delete_session_tensor(name=None)` {#delete_session_tensor} 2 | 3 | Delete the tensor by feeding a tensor handle. 4 | 5 | This is EXPERIMENTAL and subject to change. 6 | 7 | Delete the tensor of a given tensor handle. The tensor is produced 8 | in a previous run() and stored in the state of the session. 9 | 10 | ##### Args: 11 | 12 | 13 | * `name`: Optional name prefix for the return tensor. 14 | 15 | ##### Returns: 16 | 17 | A pair of graph elements. The first is a placeholder for feeding a 18 | tensor handle and the second is a deletion operation. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.device.md: -------------------------------------------------------------------------------- 1 | ### `tf.device(device_name_or_function)` {#device} 2 | 3 | Wrapper for `Graph.device()` using the default graph. 4 | 5 | See 6 | [`Graph.device()`](../../api_docs/python/framework.md#Graph.device) 7 | for more details. 8 | 9 | ##### Args: 10 | 11 | 12 | * `device_name_or_function`: The device name or function to use in 13 | the context. 14 | 15 | ##### Returns: 16 | 17 | A context manager that specifies the default device to use for newly 18 | created ops. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.errors.CancelledError.md: -------------------------------------------------------------------------------- 1 | Raised when an operation or step is cancelled. 2 | 3 | For example, a long-running operation (e.g. 4 | [`queue.enqueue()`](../../api_docs/python/io_ops.md#QueueBase.enqueue) may be 5 | cancelled by running another operation (e.g. 6 | [`queue.close(cancel_pending_enqueues=True)`](../../api_docs/python/io_ops.md#QueueBase.close), 7 | or by [closing the session](../../api_docs/python/client.md#Session.close). 8 | A step that is running such a long-running operation will fail by raising 9 | `CancelledError`. 10 | 11 | - - - 12 | 13 | #### `tf.errors.CancelledError.__init__(node_def, op, message)` {#CancelledError.__init__} 14 | 15 | Creates a `CancelledError`. 16 | 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.errors.DataLossError.md: -------------------------------------------------------------------------------- 1 | Raised when unrecoverable data loss or corruption is encountered. 2 | 3 | For example, this may be raised by running a 4 | [`tf.WholeFileReader.read()`](../../api_docs/python/io_ops.md#WholeFileReader) 5 | operation, if the file is truncated while it is being read. 6 | 7 | - - - 8 | 9 | #### `tf.errors.DataLossError.__init__(node_def, op, message)` {#DataLossError.__init__} 10 | 11 | Creates a `DataLossError`. 12 | 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.errors.DeadlineExceededError.md: -------------------------------------------------------------------------------- 1 | Raised when a deadline expires before an operation could complete. 2 | 3 | This exception is not currently used. 4 | 5 | - - - 6 | 7 | #### `tf.errors.DeadlineExceededError.__init__(node_def, op, message)` {#DeadlineExceededError.__init__} 8 | 9 | Creates a `DeadlineExceededError`. 10 | 11 | 12 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.fft.md: -------------------------------------------------------------------------------- 1 | ### `tf.fft(input, name=None)` {#fft} 2 | 3 | Compute the 1-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 vector. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. The 1D Fourier Transform of `input`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.fft2d.md: -------------------------------------------------------------------------------- 1 | ### `tf.fft2d(input, name=None)` {#fft2d} 2 | 3 | Compute the 2-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 matrix. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. The 2D Fourier Transform of `input`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.get_seed.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_seed(op_seed)` {#get_seed} 2 | 3 | Returns the local seeds an operation should use given an op-specific seed. 4 | 5 | Given operation-specific seed, `op_seed`, this helper function returns two 6 | seeds derived from graph-level and op-level seeds. Many random operations 7 | internally use the two seeds to allow user to change the seed globally for a 8 | graph, or for only specific operations. 9 | 10 | For details on how the graph-level seed interacts with op seeds, see 11 | [`set_random_seed`](../../api_docs/python/constant_op.md#set_random_seed). 12 | 13 | ##### Args: 14 | 15 | 16 | * `op_seed`: integer. 17 | 18 | ##### Returns: 19 | 20 | A tuple of two integers that should be used for the local seed of this 21 | operation. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.global_norm.md: -------------------------------------------------------------------------------- 1 | ### `tf.global_norm(t_list, name=None)` {#global_norm} 2 | 3 | Computes the global norm of multiple tensors. 4 | 5 | Given a tuple or list of tensors `t_list`, this operation returns the 6 | global norm of the elements in all tensors in `t_list`. The global norm is 7 | computed as: 8 | 9 | `global_norm = sqrt(sum([l2norm(t)**2 for t in t_list]))` 10 | 11 | Any entries in `t_list` that are of type None are ignored. 12 | 13 | ##### Args: 14 | 15 | 16 | * `t_list`: A tuple or list of mixed `Tensors`, `IndexedSlices`, or None. 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A 0-D (scalar) `Tensor` of type `float`. 22 | 23 | ##### Raises: 24 | 25 | 26 | * `TypeError`: If `t_list` is not a sequence. 27 | 28 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.ifft3d.md: -------------------------------------------------------------------------------- 1 | ### `tf.ifft3d(input, name=None)` {#ifft3d} 2 | 3 | Compute the inverse 3-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 3-D tensor. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. 14 | The inverse 3D Fourier Transform of `input`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.image.grayscale_to_rgb.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.grayscale_to_rgb(images, name=None)` {#grayscale_to_rgb} 2 | 3 | Converts one or more images from Grayscale to RGB. 4 | 5 | Outputs a tensor of the same `DType` and rank as `images`. The size of the 6 | last dimension of the output is 3, containing the RGB value of the pixels. 7 | 8 | ##### Args: 9 | 10 | 11 | * `images`: The Grayscale tensor to convert. Last dimension must be size 1. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | The converted grayscale image(s). 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.image.random_brightness.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.random_brightness(image, max_delta, seed=None)` {#random_brightness} 2 | 3 | Adjust the brightness of images by a random factor. 4 | 5 | Equivalent to `adjust_brightness()` using a `delta` randomly picked in the 6 | interval `[-max_delta, max_delta)`. 7 | 8 | ##### Args: 9 | 10 | 11 | * `image`: An image. 12 | * `max_delta`: float, must be non-negative. 13 | * `seed`: A Python integer. Used to create a random seed. See 14 | [`set_random_seed`](../../api_docs/python/constant_op.md#set_random_seed) 15 | for behavior. 16 | 17 | ##### Returns: 18 | 19 | The brightness-adjusted image. 20 | 21 | ##### Raises: 22 | 23 | 24 | * `ValueError`: if `max_delta` is negative. 25 | 26 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.image.rgb_to_grayscale.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.rgb_to_grayscale(images, name=None)` {#rgb_to_grayscale} 2 | 3 | Converts one or more images from RGB to Grayscale. 4 | 5 | Outputs a tensor of the same `DType` and rank as `images`. The size of the 6 | last dimension of the output is 1, containing the Grayscale value of the 7 | pixels. 8 | 9 | ##### Args: 10 | 11 | 12 | * `images`: The RGB tensor to convert. Last dimension must have size 3 and 13 | should contain RGB values. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | The converted grayscale image(s). 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.is_finite.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_finite(x, name=None)` {#is_finite} 2 | 3 | Returns which elements of x are finite. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `bool`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.is_nan.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_nan(x, name=None)` {#is_nan} 2 | 3 | Returns which elements of x are NaN. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `bool`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.is_non_decreasing.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_non_decreasing(x, name=None)` {#is_non_decreasing} 2 | 3 | Returns `True` if `x` is non-decreasing. 4 | 5 | Elements of `x` are compared in row-major order. The tensor `[x[0],...]` 6 | is non-decreasing if for every adjacent pair we have `x[i] <= x[i+1]`. 7 | If `x` has less than two elements, it is trivially non-decreasing. 8 | 9 | See also: `is_strictly_increasing` 10 | 11 | ##### Args: 12 | 13 | 14 | * `x`: Numeric `Tensor`. 15 | * `name`: A name for this operation (optional). Defaults to "is_non_decreasing" 16 | 17 | ##### Returns: 18 | 19 | Boolean `Tensor`, equal to `True` iff `x` is non-decreasing. 20 | 21 | ##### Raises: 22 | 23 | 24 | * `TypeError`: if `x` is not a numeric tensor. 25 | 26 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.is_numeric_tensor.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_numeric_tensor(tensor)` {#is_numeric_tensor} 2 | 3 | 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.mod.md: -------------------------------------------------------------------------------- 1 | ### `tf.mod(x, y, name=None)` {#mod} 2 | 3 | Returns element-wise remainder of division. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `int32`, `int64`, `float32`, `float64`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.mul.md: -------------------------------------------------------------------------------- 1 | ### `tf.mul(x, y, name=None)` {#mul} 2 | 3 | Returns x * y element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.name_scope.md: -------------------------------------------------------------------------------- 1 | ### `tf.name_scope(name)` {#name_scope} 2 | 3 | Wrapper for `Graph.name_scope()` using the default graph. 4 | 5 | See 6 | [`Graph.name_scope()`](../../api_docs/python/framework.md#Graph.name_scope) 7 | for more details. 8 | 9 | ##### Args: 10 | 11 | 12 | * `name`: A name for the scope. 13 | 14 | ##### Returns: 15 | 16 | A context manager that installs `name` as a new name scope in the 17 | default graph. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.nn.l2_normalize.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.l2_normalize(x, dim, epsilon=1e-12, name=None)` {#l2_normalize} 2 | 3 | Normalizes along dimension `dim` using an L2 norm. 4 | 5 | For a 1-D tensor with `dim = 0`, computes 6 | 7 | output = x / sqrt(max(sum(x**2), epsilon)) 8 | 9 | For `x` with more dimensions, independently normalizes each 1-D slice along 10 | dimension `dim`. 11 | 12 | ##### Args: 13 | 14 | 15 | * `x`: A `Tensor`. 16 | * `dim`: Dimension along which to normalize. 17 | * `epsilon`: A lower bound value for the norm. Will use `sqrt(epsilon)` as the 18 | divisor if `norm < sqrt(epsilon)`. 19 | * `name`: A name for this operation (optional). 20 | 21 | ##### Returns: 22 | 23 | A `Tensor` with the same shape as `x`. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.not_equal.md: -------------------------------------------------------------------------------- 1 | ### `tf.not_equal(x, y, name=None)` {#not_equal} 2 | 3 | Returns the truth value of (x != y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.sqrt.md: -------------------------------------------------------------------------------- 1 | ### `tf.sqrt(x, name=None)` {#sqrt} 2 | 3 | Computes square root of x element-wise. 4 | 5 | I.e., \\(y = \sqrt{x} = x^{1/2}\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.train.add_queue_runner.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.add_queue_runner(qr, collection='queue_runners')` {#add_queue_runner} 2 | 3 | Adds a `QueueRunner` to a collection in the graph. 4 | 5 | When building a complex model that uses many queues it is often difficult to 6 | gather all the queue runners that need to be run. This convenience function 7 | allows you to add a queue runner to a well known collection in the graph. 8 | 9 | The companion method `start_queue_runners()` can be used to start threads for 10 | all the collected queue runners. 11 | 12 | ##### Args: 13 | 14 | 15 | * `qr`: A `QueueRunner`. 16 | * `collection`: A `GraphKey` specifying the graph collection to add 17 | the queue runner to. Defaults to `GraphKeys.QUEUE_RUNNERS`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.train.limit_epochs.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.limit_epochs(tensor, num_epochs=None, name=None)` {#limit_epochs} 2 | 3 | Returns tensor `num_epochs` times and then raises an `OutOfRange` error. 4 | 5 | ##### Args: 6 | 7 | 8 | * `tensor`: Any `Tensor`. 9 | * `num_epochs`: A positive integer (optional). If specified, limits the number 10 | of steps the output tensor may be evaluated. 11 | * `name`: A name for the operations (optional). 12 | 13 | ##### Returns: 14 | 15 | tensor or `OutOfRange`. 16 | 17 | ##### Raises: 18 | 19 | 20 | * `ValueError`: if `num_epochs` is invalid. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard0/tf.zeros_initializer.md: -------------------------------------------------------------------------------- 1 | ### `tf.zeros_initializer(shape, dtype=tf.float32)` {#zeros_initializer} 2 | 3 | An adaptor for zeros() to match the Initializer spec. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.AggregationMethod.md: -------------------------------------------------------------------------------- 1 | A class listing aggregation methods used to combine gradients. 2 | 3 | Computing partial derivatives can require aggregating gradient 4 | contributions. This class lists the various methods that can 5 | be used to combine gradients in the graph: 6 | 7 | * `ADD_N`: All of the gradient terms are summed as part of one 8 | operation using the "AddN" op. It has the property that all 9 | gradients must be ready before any aggregation is performed. 10 | * `DEFAULT`: The system-chosen default aggregation method. 11 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.NoGradient.md: -------------------------------------------------------------------------------- 1 | ### `tf.NoGradient(op_type)` {#NoGradient} 2 | 3 | Specifies that ops of type `op_type` do not have a defined gradient. 4 | 5 | This function is only used when defining a new op type. It may be 6 | used for ops such as `tf.size()` that are not differentiable. For 7 | example: 8 | 9 | ```python 10 | tf.NoGradient("Size") 11 | ``` 12 | 13 | ##### Args: 14 | 15 | 16 | * `op_type`: The string type of an operation. This corresponds to the 17 | `OpDef.name` field for the proto that defines the operation. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `TypeError`: If `op_type` is not a string. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.Print.md: -------------------------------------------------------------------------------- 1 | ### `tf.Print(input_, data, message=None, first_n=None, summarize=None, name=None)` {#Print} 2 | 3 | Prints a list of tensors. 4 | 5 | This is an identity op with the side effect of printing `data` when 6 | evaluating. 7 | 8 | ##### Args: 9 | 10 | 11 | * `input_`: A tensor passed through this op. 12 | * `data`: A list of tensors to print out when op is evaluated. 13 | * `message`: A string, prefix of the error message. 14 | * `first_n`: Only log `first_n` number of times. Negative numbers log always; 15 | this is the default. 16 | * `summarize`: Only print this many entries of each tensor. If None, then a 17 | maximum of 3 elements are printed per input tensor. 18 | * `name`: A name for the operation (optional). 19 | 20 | ##### Returns: 21 | 22 | Same tensor as `input_`. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.Variable.from_proto.md: -------------------------------------------------------------------------------- 1 | #### `tf.Variable.from_proto(variable_def)` {#Variable.from_proto} 2 | 3 | Returns a `Variable` object created from `variable_def`. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.all_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.all_variables()` {#all_variables} 2 | 3 | Returns all variables that must be saved/restored. 4 | 5 | The `Variable()` constructor automatically adds new variables to the graph 6 | collection `GraphKeys.VARIABLES`. This convenience function returns the 7 | contents of that collection. 8 | 9 | ##### Returns: 10 | 11 | A list of `Variable` objects. 12 | 13 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.batch_fft.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_fft(input, name=None)` {#batch_fft} 2 | 3 | Compute the 1-dimensional discrete Fourier Transform over the inner-most 4 | 5 | dimension of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 17 | dimension of `input` is replaced with its 1D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.constant_initializer.md: -------------------------------------------------------------------------------- 1 | ### `tf.constant_initializer(value=0.0, dtype=tf.float32)` {#constant_initializer} 2 | 3 | Returns an initializer that generates tensors with a single value. 4 | 5 | ##### Args: 6 | 7 | 8 | * `value`: A Python scalar. All elements of the initialized variable 9 | will be set to this value. 10 | * `dtype`: The data type. Only floating point types are supported. 11 | 12 | ##### Returns: 13 | 14 | An initializer that generates tensors with a single value. 15 | 16 | ##### Raises: 17 | 18 | 19 | * `ValueError`: if `dtype` is not a floating point type. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.framework.get_local_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_local_variables(scope=None, suffix=None)` {#get_local_variables} 2 | 3 | Gets the list of model variables, filtered by scope and/or suffix. 4 | 5 | ##### Args: 6 | 7 | 8 | * `scope`: an optional scope for filtering the variables to return. 9 | * `suffix`: an optional suffix for filtering the variables to return. 10 | 11 | ##### Returns: 12 | 13 | a list of variables in colelction with scope and suffix. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.framework.get_variables_by_name.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_variables_by_name(given_name, scope=None)` {#get_variables_by_name} 2 | 3 | Gets the list of variables that were given that name. 4 | 5 | ##### Args: 6 | 7 | 8 | * `given_name`: name given to the variable without any scope. 9 | * `scope`: an optional scope for filtering the variables to return. 10 | 11 | ##### Returns: 12 | 13 | a copied list of variables with the given name and scope. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.layers.summarize_tensors.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.summarize_tensors(tensors, summarizer=summarize_tensor)` {#summarize_tensors} 2 | 3 | Summarize a set of tensors. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.contrib.learn.extract_pandas_data.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.extract_pandas_data(data)` {#extract_pandas_data} 2 | 3 | Extract data from pandas.DataFrame for predictors. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.erf.md: -------------------------------------------------------------------------------- 1 | ### `tf.erf(x, name=None)` {#erf} 2 | 3 | Computes the Gauss error function of `x` element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.errors.AlreadyExistsError.md: -------------------------------------------------------------------------------- 1 | Raised when an entity that we attempted to create already exists. 2 | 3 | For example, running an operation that saves a file 4 | (e.g. [`tf.train.Saver.save()`](../../api_docs/python/train.md#Saver.save)) 5 | could potentially raise this exception if an explicit filename for an 6 | existing file was passed. 7 | 8 | - - - 9 | 10 | #### `tf.errors.AlreadyExistsError.__init__(node_def, op, message)` {#AlreadyExistsError.__init__} 11 | 12 | Creates an `AlreadyExistsError`. 13 | 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.get_default_session.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_default_session()` {#get_default_session} 2 | 3 | Returns the default session for the current thread. 4 | 5 | The returned `Session` will be the innermost session on which a 6 | `Session` or `Session.as_default()` context has been entered. 7 | 8 | NOTE: The default session is a property of the current thread. If you 9 | create a new thread, and wish to use the default session in that 10 | thread, you must explicitly add a `with sess.as_default():` in that 11 | thread's function. 12 | 13 | ##### Returns: 14 | 15 | The default `Session` being used in the current thread. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.greater_equal.md: -------------------------------------------------------------------------------- 1 | ### `tf.greater_equal(x, y, name=None)` {#greater_equal} 2 | 3 | Returns the truth value of (x >= y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.igammac.md: -------------------------------------------------------------------------------- 1 | ### `tf.igammac(a, x, name=None)` {#igammac} 2 | 3 | Compute the upper regularized incomplete Gamma function `Q(a, x)`. 4 | 5 | The upper regularized incomplete Gamma function is defined as: 6 | 7 | ``` 8 | Q(a, x) = Gamma(a, x) / Gamma(x) = 1 - P(a, x) 9 | ``` 10 | where 11 | ``` 12 | Gamma(a, x) = int_{x}^{\infty} t^{a-1} exp(-t) dt 13 | ``` 14 | is the upper incomplete Gama function. 15 | 16 | Note, above `P(a, x)` (`Igamma`) is the lower regularized complete 17 | Gamma function. 18 | 19 | ##### Args: 20 | 21 | 22 | * `a`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 23 | * `x`: A `Tensor`. Must have the same type as `a`. 24 | * `name`: A name for the operation (optional). 25 | 26 | ##### Returns: 27 | 28 | A `Tensor`. Has the same type as `a`. 29 | 30 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.image.central_crop.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.central_crop(image, central_fraction)` {#central_crop} 2 | 3 | Crop the central region of the image. 4 | 5 | Remove the outer parts of an image but retain the central region of the image 6 | along each dimension. If we specify central_fraction = 0.5, this function 7 | returns the region marked with "X" in the below diagram. 8 | 9 | -------- 10 | | | 11 | | XXXX | 12 | | XXXX | 13 | | | where "X" is the central 50% of the image. 14 | -------- 15 | 16 | ##### Args: 17 | 18 | 19 | * `image`: 3-D float Tensor of shape [height, width, depth] 20 | * `central_fraction`: float (0, 1], fraction of size to crop 21 | 22 | ##### Raises: 23 | 24 | 25 | * `ValueError`: if central_crop_fraction is not within (0, 1]. 26 | 27 | ##### Returns: 28 | 29 | 3-D float Tensor 30 | 31 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.is_strictly_increasing.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_strictly_increasing(x, name=None)` {#is_strictly_increasing} 2 | 3 | Returns `True` if `x` is strictly increasing. 4 | 5 | Elements of `x` are compared in row-major order. The tensor `[x[0],...]` 6 | is strictly increasing if for every adjacent pair we have `x[i] < x[i+1]`. 7 | If `x` has less than two elements, it is trivially strictly increasing. 8 | 9 | See also: `is_non_decreasing` 10 | 11 | ##### Args: 12 | 13 | 14 | * `x`: Numeric `Tensor`. 15 | * `name`: A name for this operation (optional). 16 | Defaults to "is_strictly_increasing" 17 | 18 | ##### Returns: 19 | 20 | Boolean `Tensor`, equal to `True` iff `x` is strictly increasing. 21 | 22 | ##### Raises: 23 | 24 | 25 | * `TypeError`: if `x` is not a numeric tensor. 26 | 27 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.matching_files.md: -------------------------------------------------------------------------------- 1 | ### `tf.matching_files(pattern, name=None)` {#matching_files} 2 | 3 | Returns the set of files matching a pattern. 4 | 5 | Note that this routine only supports wildcard characters in the 6 | basename portion of the pattern, not in the directory portion. 7 | 8 | ##### Args: 9 | 10 | 11 | * `pattern`: A `Tensor` of type `string`. A (scalar) shell wildcard pattern. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A `Tensor` of type `string`. A vector of matching filenames. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.merge_all_summaries.md: -------------------------------------------------------------------------------- 1 | ### `tf.merge_all_summaries(key='summaries')` {#merge_all_summaries} 2 | 3 | Merges all summaries collected in the default graph. 4 | 5 | ##### Args: 6 | 7 | 8 | * `key`: `GraphKey` used to collect the summaries. Defaults to 9 | `GraphKeys.SUMMARIES`. 10 | 11 | ##### Returns: 12 | 13 | If no summaries were collected, returns None. Otherwise returns a scalar 14 | `Tensor` of type `string` containing the serialized `Summary` protocol 15 | buffer resulting from the merging. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.read_file.md: -------------------------------------------------------------------------------- 1 | ### `tf.read_file(filename, name=None)` {#read_file} 2 | 3 | Reads and outputs the entire contents of the input filename. 4 | 5 | ##### Args: 6 | 7 | 8 | * `filename`: A `Tensor` of type `string`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `string`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.test.get_temp_dir.md: -------------------------------------------------------------------------------- 1 | ### `tf.test.get_temp_dir()` {#get_temp_dir} 2 | 3 | Returns a temporary directory for use during tests. 4 | 5 | There is no need to delete the directory after the test. 6 | 7 | ##### Returns: 8 | 9 | The temporary directory. 10 | 11 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.train.GradientDescentOptimizer.md: -------------------------------------------------------------------------------- 1 | Optimizer that implements the gradient descent algorithm. 2 | 3 | - - - 4 | 5 | #### `tf.train.GradientDescentOptimizer.__init__(learning_rate, use_locking=False, name='GradientDescent')` {#GradientDescentOptimizer.__init__} 6 | 7 | Construct a new gradient descent optimizer. 8 | 9 | ##### Args: 10 | 11 | 12 | * `learning_rate`: A Tensor or a floating point value. The learning 13 | rate to use. 14 | * `use_locking`: If True use locks for update operations. 15 | * `name`: Optional name prefix for the operations created when applying 16 | gradients. Defaults to "GradientDescent". 17 | 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.train.Server.create_local_server.md: -------------------------------------------------------------------------------- 1 | #### `tf.train.Server.create_local_server(start=True)` {#Server.create_local_server} 2 | 3 | Creates a new single-process cluster running on the local host. 4 | 5 | This method is a convenience wrapper for creating a 6 | `tf.train.Server` with a `tf.train.ServerDef` that specifies a 7 | single-process cluster containing a single task in a job called 8 | `"local"`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `start`: (Optional.) Boolean, indicating whether to start the server after 14 | creating it. Defaults to `True`. 15 | 16 | ##### Returns: 17 | 18 | A local `tf.train.Server`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.train.generate_checkpoint_state_proto.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.generate_checkpoint_state_proto(save_dir, model_checkpoint_path, all_model_checkpoint_paths=None)` {#generate_checkpoint_state_proto} 2 | 3 | Generates a checkpoint state proto. 4 | 5 | ##### Args: 6 | 7 | 8 | * `save_dir`: Directory where the model was saved. 9 | * `model_checkpoint_path`: The checkpoint file. 10 | * `all_model_checkpoint_paths`: List of strings. Paths to all not-yet-deleted 11 | checkpoints, sorted from oldest to newest. If this is a non-empty list, 12 | the last element must be equal to model_checkpoint_path. These paths 13 | are also saved in the CheckpointState proto. 14 | 15 | ##### Returns: 16 | 17 | CheckpointState proto with model_checkpoint_path and 18 | all_model_checkpoint_paths updated to either absolute paths or 19 | relative paths to the current save_dir. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard1/tf.zeros.md: -------------------------------------------------------------------------------- 1 | ### `tf.zeros(shape, dtype=tf.float32, name=None)` {#zeros} 2 | 3 | Creates a tensor with all elements set to zero. 4 | 5 | This operation returns a tensor of type `dtype` with shape `shape` and 6 | all elements set to zero. 7 | 8 | For example: 9 | 10 | ```python 11 | tf.zeros([3, 4], int32) ==> [[0, 0, 0, 0], [0, 0, 0, 0], [0, 0, 0, 0]] 12 | ``` 13 | 14 | ##### Args: 15 | 16 | 17 | * `shape`: Either a list of integers, or a 1-D `Tensor` of type `int32`. 18 | * `dtype`: The type of an element in the resulting `Tensor`. 19 | * `name`: A name for the operation (optional). 20 | 21 | ##### Returns: 22 | 23 | A `Tensor` with all elements set to zero. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.batch_matrix_determinant.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_matrix_determinant(input, name=None)` {#batch_matrix_determinant} 2 | 3 | Calculates the determinants for a batch of square matrices. 4 | 5 | The input is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions 6 | form square matrices. The output is a 1-D tensor containing the determinants 7 | for all input submatrices `[..., :, :]`. 8 | 9 | ##### Args: 10 | 11 | 12 | * `input`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 13 | Shape is `[..., M, M]`. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | A `Tensor`. Has the same type as `input`. Shape is `[...]`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.copy_graph.get_copied_op.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.copy_graph.get_copied_op(org_instance, graph, scope='')` {#get_copied_op} 2 | 3 | Given an `Operation` instance from some `Graph`, returns 4 | its namesake from `graph`, under the specified scope 5 | (default `""`). 6 | 7 | If a copy of `org_instance` is present in `graph` under the given 8 | `scope`, it will be returned. 9 | 10 | Args: 11 | org_instance: An `Operation` from some `Graph`. 12 | graph: The `Graph` to be searched for a copr of `org_instance`. 13 | scope: The scope `org_instance` is present in. 14 | 15 | ##### Returns: 16 | 17 | The `Operation` copy from `graph`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.framework.assert_global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.assert_global_step(global_step_tensor)` {#assert_global_step} 2 | 3 | Asserts `global_step_tensor` is a scalar int `Variable` or `Tensor`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `global_step_tensor`: `Tensor` to test. 9 | 10 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.framework.assert_scalar_int.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.assert_scalar_int(tensor)` {#assert_scalar_int} 2 | 3 | Assert `tensor` is 0-D, of type `tf.int32` or `tf.int64`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `tensor`: Tensor to test. 9 | 10 | ##### Returns: 11 | 12 | `tensor`, for chaining. 13 | 14 | ##### Raises: 15 | 16 | 17 | * `ValueError`: if `tensor` is not 0-D, of type `tf.int32` or `tf.int64`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.framework.get_unique_variable.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_unique_variable(var_op_name)` {#get_unique_variable} 2 | 3 | Gets the variable uniquely identified by that var_op_name. 4 | 5 | ##### Args: 6 | 7 | 8 | * `var_op_name`: the full name of the variable op, including the scope. 9 | 10 | ##### Returns: 11 | 12 | a tensorflow variable. 13 | 14 | ##### Raises: 15 | 16 | 17 | * `ValueError`: if no variable uniquely identified by the name exists. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.framework.get_variables_to_restore.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_variables_to_restore(include=None, exclude=None)` {#get_variables_to_restore} 2 | 3 | Gets the list of the variables to restore. 4 | 5 | ##### Args: 6 | 7 | 8 | * `include`: an optional list/tuple of scope strings for filtering which 9 | variables from the VARIABLES collection to include. None would include all 10 | the variables. 11 | * `exclude`: an optional list/tuple of scope strings for filtering which 12 | variables from the VARIABLES collection to exclude. None it would not 13 | exclude any. 14 | 15 | ##### Returns: 16 | 17 | a list of variables to restore. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `TypeError`: include or exclude is provided but is not a list or a tuple. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.framework.with_same_shape.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.with_same_shape(expected_tensor, tensor)` {#with_same_shape} 2 | 3 | Assert tensors are the same shape, from the same graph. 4 | 5 | ##### Args: 6 | 7 | 8 | * `expected_tensor`: Tensor with expected shape. 9 | * `tensor`: Tensor of actual values. 10 | 11 | ##### Returns: 12 | 13 | Tuple of (actual_tensor, label_tensor), possibly with assert ops added. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.contrib.learn.ModeKeys.md: -------------------------------------------------------------------------------- 1 | Standard names for model modes. 2 | 3 | The following standard keys are defined: 4 | 5 | * `TRAIN`: training mode. 6 | * `EVAL`: evaluation mode. 7 | * `INFER`: inference mode. 8 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.digamma.md: -------------------------------------------------------------------------------- 1 | ### `tf.digamma(x, name=None)` {#digamma} 2 | 3 | Computes Psi, the derivative of Lgamma (the log of the absolute value of 4 | 5 | `Gamma(x)`), element-wise. 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.errors.ResourceExhaustedError.md: -------------------------------------------------------------------------------- 1 | Some resource has been exhausted. 2 | 3 | For example, this error might be raised if a per-user quota is 4 | exhausted, or perhaps the entire file system is out of space. 5 | 6 | - - - 7 | 8 | #### `tf.errors.ResourceExhaustedError.__init__(node_def, op, message)` {#ResourceExhaustedError.__init__} 9 | 10 | Creates a `ResourceExhaustedError`. 11 | 12 | 13 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.get_variable_scope.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_variable_scope()` {#get_variable_scope} 2 | 3 | Returns the current variable scope. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.identity.md: -------------------------------------------------------------------------------- 1 | ### `tf.identity(input, name=None)` {#identity} 2 | 3 | Return a tensor with the same shape and contents as the input tensor or value. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `input`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.imag.md: -------------------------------------------------------------------------------- 1 | ### `tf.imag(input, name=None)` {#imag} 2 | 3 | Returns the imaginary part of a complex number. 4 | 5 | Given a tensor `input` of complex numbers, this operation returns a tensor of 6 | type `float` or `double` that is the imaginary part of each element in 7 | `input`. All elements in `input` must be complex numbers of the form \(a + 8 | bj\), where *a* is the real part and *b* is the imaginary part returned by 9 | this operation. 10 | 11 | For example: 12 | 13 | ``` 14 | # tensor 'input' is [-2.25 + 4.75j, 3.25 + 5.75j] 15 | tf.imag(input) ==> [4.75, 5.75] 16 | ``` 17 | 18 | ##### Args: 19 | 20 | 21 | * `input`: A `Tensor`. Must be one of the following types: `complex64`, `complex128`. 22 | * `name`: A name for the operation (optional). 23 | 24 | ##### Returns: 25 | 26 | A `Tensor` of type `float` or `double`. 27 | 28 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.initialize_all_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.initialize_all_variables()` {#initialize_all_variables} 2 | 3 | Returns an Op that initializes all variables. 4 | 5 | This is just a shortcut for `initialize_variables(all_variables())` 6 | 7 | ##### Returns: 8 | 9 | An Op that initializes all variables in the graph. 10 | 11 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.initialize_local_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.initialize_local_variables()` {#initialize_local_variables} 2 | 3 | Returns an Op that initializes all local variables. 4 | 5 | This is just a shortcut for `initialize_variables(local_variables())` 6 | 7 | ##### Returns: 8 | 9 | An Op that initializes all local variables in the graph. 10 | 11 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.is_variable_initialized.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_variable_initialized(variable)` {#is_variable_initialized} 2 | 3 | Tests if a variable has been initialized. 4 | 5 | ##### Args: 6 | 7 | 8 | * `variable`: A `Variable`. 9 | 10 | ##### Returns: 11 | 12 | Returns a scalar boolean Tensor, `True` if the variable has been 13 | initialized, `False` otherwise. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.minimum.md: -------------------------------------------------------------------------------- 1 | ### `tf.minimum(x, y, name=None)` {#minimum} 2 | 3 | Returns the min of x and y (i.e. x < y ? x : y) element-wise, broadcasts. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.neg.md: -------------------------------------------------------------------------------- 1 | ### `tf.neg(x, name=None)` {#neg} 2 | 3 | Computes numerical negative value element-wise. 4 | 5 | I.e., \\(y = -x\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.nn.relu.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.relu(features, name=None)` {#relu} 2 | 3 | Computes rectified linear: `max(features, 0)`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `features`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `features`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.real.md: -------------------------------------------------------------------------------- 1 | ### `tf.real(input, name=None)` {#real} 2 | 3 | Returns the real part of a complex number. 4 | 5 | Given a tensor `input` of complex numbers, this operation returns a tensor of 6 | type `float` or `double` that is the real part of each element in `input`. 7 | All elements in `input` must be complex numbers of the form \(a + bj\), 8 | where *a* is the real part returned by this operation and *b* is the 9 | imaginary part. 10 | 11 | For example: 12 | 13 | ``` 14 | # tensor 'input' is [-2.25 + 4.75j, 3.25 + 5.75j] 15 | tf.real(input) ==> [-2.25, 3.25] 16 | ``` 17 | 18 | ##### Args: 19 | 20 | 21 | * `input`: A `Tensor`. Must be one of the following types: `complex64`, 22 | `complex128`. 23 | * `name`: A name for the operation (optional). 24 | 25 | ##### Returns: 26 | 27 | A `Tensor` of type `float` or `double`. 28 | 29 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.report_uninitialized_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.report_uninitialized_variables(var_list=None, name='report_uninitialized_variables')` {#report_uninitialized_variables} 2 | 3 | Adds ops to list the names of uninitialized variables. 4 | 5 | When run, it returns a 1-D tensor containing the names of uninitialized 6 | variables if there are any, or an empty array if there are none. 7 | 8 | ##### Args: 9 | 10 | 11 | * `var_list`: List of `Variable` objects to check. Defaults to the 12 | value of `all_variables() + local_variables()` 13 | * `name`: Optional name of the `Operation`. 14 | 15 | ##### Returns: 16 | 17 | A 1-D tensor containing names of the unintialized variables, or an empty 1-D 18 | tensor if there are no variables or no uninitialized variables. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.shape_n.md: -------------------------------------------------------------------------------- 1 | ### `tf.shape_n(input, name=None)` {#shape_n} 2 | 3 | Returns shape of tensors. 4 | 5 | This operation returns N 1-D integer tensors representing shape of `input[i]s`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A list of at least 1 `Tensor` objects of the same type. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A list with the same number of `Tensor` objects as `input` of `Tensor` objects of type `int32`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.sin.md: -------------------------------------------------------------------------------- 1 | ### `tf.sin(x, name=None)` {#sin} 2 | 3 | Computes sin of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.test.assert_equal_graph_def.md: -------------------------------------------------------------------------------- 1 | ### `tf.test.assert_equal_graph_def(actual, expected)` {#assert_equal_graph_def} 2 | 3 | Asserts that two `GraphDef`s are (mostly) the same. 4 | 5 | Compares two `GraphDef` protos for equality, ignoring versions and ordering of 6 | nodes, attrs, and control inputs. Node names are used to match up nodes 7 | between the graphs, so the naming of nodes must be consistent. 8 | 9 | ##### Args: 10 | 11 | 12 | * `actual`: The `GraphDef` we have. 13 | * `expected`: The `GraphDef` we expected. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `AssertionError`: If the `GraphDef`s do not match. 19 | * `TypeError`: If either argument is not a `GraphDef`. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.test.is_built_with_cuda.md: -------------------------------------------------------------------------------- 1 | ### `tf.test.is_built_with_cuda()` {#is_built_with_cuda} 2 | 3 | Returns whether TensorFlow was built with CUDA (GPU) support. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.to_int32.md: -------------------------------------------------------------------------------- 1 | ### `tf.to_int32(x, name='ToInt32')` {#to_int32} 2 | 3 | Casts a tensor to type `int32`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` or `SparseTensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` or `SparseTensor` with same shape as `x` with type `int32`. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `TypeError`: If `x` cannot be cast to the `int32`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.train.Saver.from_proto.md: -------------------------------------------------------------------------------- 1 | #### `tf.train.Saver.from_proto(saver_def)` {#Saver.from_proto} 2 | 3 | Returns a `Saver` object created from `saver_def`. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.train.global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.global_step(sess, global_step_tensor)` {#global_step} 2 | 3 | Small helper to get the global step. 4 | 5 | ```python 6 | # Creates a variable to hold the global_step. 7 | global_step_tensor = tf.Variable(10, trainable=False, name='global_step') 8 | # Creates a session. 9 | sess = tf.Session() 10 | # Initializes the variable. 11 | sess.run(global_step_tensor.initializer) 12 | print('global_step: %s' % tf.train.global_step(sess, global_step_tensor)) 13 | 14 | global_step: 10 15 | ``` 16 | 17 | ##### Args: 18 | 19 | 20 | * `sess`: A TensorFlow `Session` object. 21 | * `global_step_tensor`: `Tensor` or the `name` of the operation that contains 22 | the global step. 23 | 24 | ##### Returns: 25 | 26 | The global step value. 27 | 28 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard2/tf.train.latest_checkpoint.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.latest_checkpoint(checkpoint_dir, latest_filename=None)` {#latest_checkpoint} 2 | 3 | Finds the filename of latest saved checkpoint file. 4 | 5 | ##### Args: 6 | 7 | 8 | * `checkpoint_dir`: Directory where the variables were saved. 9 | * `latest_filename`: Optional name for the protocol buffer file that 10 | contains the list of most recent checkpoint filenames. 11 | See the corresponding argument to `Saver.save()`. 12 | 13 | ##### Returns: 14 | 15 | The full path to the latest checkpoint or `None` if no checkpoint was found. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.assert_type.md: -------------------------------------------------------------------------------- 1 | ### `tf.assert_type(tensor, tf_type)` {#assert_type} 2 | 3 | Asserts that the given `Tensor` is of the specified type. 4 | 5 | ##### Args: 6 | 7 | 8 | * `tensor`: A tensorflow `Tensor`. 9 | * `tf_type`: A tensorflow type (dtypes.float32, tf.int64, dtypes.bool, etc). 10 | 11 | ##### Raises: 12 | 13 | 14 | * `ValueError`: If the tensors data type doesn't match tf_type. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.batch_ifft3d.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_ifft3d(input, name=None)` {#batch_ifft3d} 2 | 3 | Compute the inverse 3-dimensional discrete Fourier Transform over the inner-most 4 | 5 | 3 dimensions of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 3 17 | dimensions of `input` are replaced with their inverse 3D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.batch_self_adjoint_eig.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_self_adjoint_eig(input, name=None)` {#batch_self_adjoint_eig} 2 | 3 | Calculates the Eigen Decomposition of a batch of square self-adjoint matrices. 4 | 5 | The input is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions 6 | form square matrices, with the same constraints as the single matrix 7 | SelfAdjointEig. 8 | 9 | The result is a '[..., M+1, M] matrix with [..., 0,:] containing the 10 | eigenvalues, and subsequent [...,1:, :] containing the eigenvectors. 11 | 12 | ##### Args: 13 | 14 | 15 | * `input`: A `Tensor`. Must be one of the following types: `float64`, `float32`. 16 | Shape is `[..., M, M]`. 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A `Tensor`. Has the same type as `input`. Shape is `[..., M+1, M]`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.ceil.md: -------------------------------------------------------------------------------- 1 | ### `tf.ceil(x, name=None)` {#ceil} 2 | 3 | Returns element-wise smallest integer in not less than x. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.check_numerics.md: -------------------------------------------------------------------------------- 1 | ### `tf.check_numerics(tensor, message, name=None)` {#check_numerics} 2 | 3 | Checks a tensor for NaN and Inf values. 4 | 5 | When run, reports an `InvalidArgument` error if `tensor` has any values 6 | that are not a number (NaN) or infinity (Inf). Otherwise, passes `tensor` as-is. 7 | 8 | ##### Args: 9 | 10 | 11 | * `tensor`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 12 | * `message`: A `string`. Prefix of the error message. 13 | * `name`: A name for the operation (optional). 14 | 15 | ##### Returns: 16 | 17 | A `Tensor`. Has the same type as `tensor`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.complex_abs.md: -------------------------------------------------------------------------------- 1 | ### `tf.complex_abs(x, name=None)` {#complex_abs} 2 | 3 | Computes the complex absolute value of a tensor. 4 | 5 | Given a tensor `x` of complex numbers, this operation returns a tensor of type 6 | `float` or `double` that is the absolute value of each element in `x`. All 7 | elements in `x` must be complex numbers of the form \\(a + bj\\). The 8 | absolute value is computed as \\( \sqrt{a^2 + b^2}\\). 9 | 10 | For example: 11 | 12 | ``` 13 | # tensor 'x' is [[-2.25 + 4.75j], [-3.25 + 5.75j]] 14 | tf.complex_abs(x) ==> [5.25594902, 6.60492229] 15 | ``` 16 | 17 | ##### Args: 18 | 19 | 20 | * `x`: A `Tensor` of type `complex64` or `complex128`. 21 | * `name`: A name for the operation (optional). 22 | 23 | ##### Returns: 24 | 25 | A `Tensor` of type `float32` or `float64`. 26 | 27 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.ffmpeg.encode_audio.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.ffmpeg.encode_audio(audio, file_format=None, samples_per_second=None)` {#encode_audio} 2 | 3 | Creates an op that encodes an audio file using sampled audio from a tensor. 4 | 5 | ##### Args: 6 | 7 | 8 | * `audio`: A rank 2 tensor that has time along dimension 0 and channels along 9 | dimension 1. Dimension 0 is `samples_per_second * length` long in 10 | seconds. 11 | * `file_format`: The type of file to encode. "wav" is the only supported format. 12 | * `samples_per_second`: The number of samples in the audio tensor per second of 13 | audio. 14 | 15 | ##### Returns: 16 | 17 | A scalar tensor that contains the encoded audio in the specified file 18 | format. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.framework.add_model_variable.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.add_model_variable(var)` {#add_model_variable} 2 | 3 | Adds a variable to the MODEL_VARIABLES collection. 4 | 5 | ##### Args: 6 | 7 | 8 | * `var`: a variable. 9 | 10 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.framework.get_model_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_model_variables(scope=None, suffix=None)` {#get_model_variables} 2 | 3 | Gets the list of model variables, filtered by scope and/or suffix. 4 | 5 | ##### Args: 6 | 7 | 8 | * `scope`: an optional scope for filtering the variables to return. 9 | * `suffix`: an optional suffix for filtering the variables to return. 10 | 11 | ##### Returns: 12 | 13 | a list of variables in colelction with scope and suffix. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.framework.local_variable.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.local_variable(initial_value, validate_shape=True, name=None)` {#local_variable} 2 | 3 | Create variable and add it to `GraphKeys.LOCAL_VARIABLES` collection. 4 | 5 | ##### Args: 6 | 7 | 8 | * `initial_value`: See variables.Variable.__init__. 9 | * `validate_shape`: See variables.Variable.__init__. 10 | * `name`: See variables.Variable.__init__. 11 | 12 | ##### Returns: 13 | 14 | New variable. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.layers.summarize_activation.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.summarize_activation(op)` {#summarize_activation} 2 | 3 | Summarize an activation. 4 | 5 | This applies the given activation and adds useful summaries specific to the 6 | activation. 7 | 8 | ##### Args: 9 | 10 | 11 | * `op`: The tensor to summarize (assumed to be a layer activation). 12 | 13 | ##### Returns: 14 | 15 | The summary op created to summarize `op`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.learn.extract_dask_data.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.extract_dask_data(data)` {#extract_dask_data} 2 | 3 | Extract data from dask.Series or dask.DataFrame for predictors. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.losses.add_loss.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.losses.add_loss(loss)` {#add_loss} 2 | 3 | Adds a externally defined loss to collection of losses. 4 | 5 | ##### Args: 6 | 7 | 8 | * `loss`: A loss `Tensor`. 9 | 10 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.contrib.losses.get_regularization_losses.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.losses.get_regularization_losses(scope=None)` {#get_regularization_losses} 2 | 3 | Gets the regularization losses. 4 | 5 | ##### Args: 6 | 7 | 8 | * `scope`: an optional scope for filtering the losses to return. 9 | 10 | ##### Returns: 11 | 12 | A list of loss variables. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.control_dependencies.md: -------------------------------------------------------------------------------- 1 | ### `tf.control_dependencies(control_inputs)` {#control_dependencies} 2 | 3 | Wrapper for `Graph.control_dependencies()` using the default graph. 4 | 5 | See [`Graph.control_dependencies()`](../../api_docs/python/framework.md#Graph.control_dependencies) 6 | for more details. 7 | 8 | ##### Args: 9 | 10 | 11 | * `control_inputs`: A list of `Operation` or `Tensor` objects which 12 | must be executed or computed before running the operations 13 | defined in the context. Can also be `None` to clear the control 14 | dependencies. 15 | 16 | ##### Returns: 17 | 18 | A context manager that specifies control dependencies for all 19 | operations constructed within the context. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.errors.OutOfRangeError.md: -------------------------------------------------------------------------------- 1 | Raised when an operation iterates past the valid input range. 2 | 3 | This exception is raised in "end-of-file" conditions, such as when a 4 | [`queue.dequeue()`](../../api_docs/python/io_ops.md#QueueBase.dequeue) 5 | operation is blocked on an empty queue, and a 6 | [`queue.close()`](../../api_docs/python/io_ops.md#QueueBase.close) 7 | operation executes. 8 | 9 | - - - 10 | 11 | #### `tf.errors.OutOfRangeError.__init__(node_def, op, message)` {#OutOfRangeError.__init__} 12 | 13 | Creates an `OutOfRangeError`. 14 | 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.errors.UnauthenticatedError.md: -------------------------------------------------------------------------------- 1 | The request does not have valid authentication credentials. 2 | 3 | This exception is not currently used. 4 | 5 | - - - 6 | 7 | #### `tf.errors.UnauthenticatedError.__init__(node_def, op, message)` {#UnauthenticatedError.__init__} 8 | 9 | Creates an `UnauthenticatedError`. 10 | 11 | 12 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.exp.md: -------------------------------------------------------------------------------- 1 | ### `tf.exp(x, name=None)` {#exp} 2 | 3 | Computes exponential of x element-wise. \\(y = e^x\\). 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.group.md: -------------------------------------------------------------------------------- 1 | ### `tf.group(*inputs, **kwargs)` {#group} 2 | 3 | Create an op that groups multiple operations. 4 | 5 | When this op finishes, all ops in `input` have finished. This op has no 6 | output. 7 | 8 | See also `tuple` and `with_dependencies`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `*inputs`: Zero or more tensors to group. 14 | * `**kwargs`: Optional parameters to pass when constructing the NodeDef. 15 | * `name`: A name for this operation (optional). 16 | 17 | ##### Returns: 18 | 19 | An Operation that executes all its inputs. 20 | 21 | ##### Raises: 22 | 23 | 24 | * `ValueError`: If an unknown keyword argument is provided. 25 | 26 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.ifft2d.md: -------------------------------------------------------------------------------- 1 | ### `tf.ifft2d(input, name=None)` {#ifft2d} 2 | 3 | Compute the inverse 2-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 matrix. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. 14 | The inverse 2D Fourier Transform of `input`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.image.random_contrast.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.random_contrast(image, lower, upper, seed=None)` {#random_contrast} 2 | 3 | Adjust the contrast of an image by a random factor. 4 | 5 | Equivalent to `adjust_contrast()` but uses a `contrast_factor` randomly 6 | picked in the interval `[lower, upper]`. 7 | 8 | ##### Args: 9 | 10 | 11 | * `image`: An image tensor with 3 or more dimensions. 12 | * `lower`: float. Lower bound for the random contrast factor. 13 | * `upper`: float. Upper bound for the random contrast factor. 14 | * `seed`: A Python integer. Used to create a random seed. See 15 | [`set_random_seed`](../../api_docs/python/constant_op.md#set_random_seed) 16 | for behavior. 17 | 18 | ##### Returns: 19 | 20 | The contrast-adjusted tensor. 21 | 22 | ##### Raises: 23 | 24 | 25 | * `ValueError`: if `upper <= lower` or if `lower < 0`. 26 | 27 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.local_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.local_variables()` {#local_variables} 2 | 3 | Returns all variables created with collection=[LOCAL_VARIABLES]. 4 | 5 | ##### Returns: 6 | 7 | A list of local Variable objects. 8 | 9 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.logical_xor.md: -------------------------------------------------------------------------------- 1 | ### `tf.logical_xor(x, y, name='LogicalXor')` {#logical_xor} 2 | 3 | x ^ y = (x | y) & ~(x & y). 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.log_softmax.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.log_softmax(logits, name=None)` {#log_softmax} 2 | 3 | Computes log softmax activations. 4 | 5 | For each batch `i` and class `j` we have 6 | 7 | logsoftmax[i, j] = logits[i, j] - log(sum(exp(logits[i]))) 8 | 9 | ##### Args: 10 | 11 | 12 | * `logits`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 13 | 2-D with shape `[batch_size, num_classes]`. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | A `Tensor`. Has the same type as `logits`. Same shape as `logits`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.nn.softsign.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.softsign(features, name=None)` {#softsign} 2 | 3 | Computes softsign: `features / (abs(features) + 1)`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `features`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `features`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.pack.md: -------------------------------------------------------------------------------- 1 | ### `tf.pack(values, name='pack')` {#pack} 2 | 3 | Packs a list of rank-`R` tensors into one rank-`(R+1)` tensor. 4 | 5 | Packs tensors in `values` into a tensor with rank one higher than each tensor 6 | in `values` and shape `[len(values)] + values[0].shape`. The output satisfies 7 | `output[i, ...] = values[i][...]`. 8 | 9 | This is the opposite of unpack. The numpy equivalent is 10 | 11 | tf.pack([x, y, z]) = np.asarray([x, y, z]) 12 | 13 | ##### Args: 14 | 15 | 16 | * `values`: A list of `Tensor` objects with the same shape and type. 17 | * `name`: A name for this operation (optional). 18 | 19 | ##### Returns: 20 | 21 | 22 | * `output`: A packed `Tensor` with the same type as `values`. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.scalar_summary.md: -------------------------------------------------------------------------------- 1 | ### `tf.scalar_summary(tags, values, collections=None, name=None)` {#scalar_summary} 2 | 3 | Outputs a `Summary` protocol buffer with scalar values. 4 | 5 | The input `tags` and `values` must have the same shape. The generated 6 | summary has a summary value for each tag-value pair in `tags` and `values`. 7 | 8 | ##### Args: 9 | 10 | 11 | * `tags`: A `string` `Tensor`. Tags for the summaries. 12 | * `values`: A real numeric Tensor. Values for the summaries. 13 | * `collections`: Optional list of graph collections keys. The new summary op is 14 | added to these collections. Defaults to `[GraphKeys.SUMMARIES]`. 15 | * `name`: A name for the operation (optional). 16 | 17 | ##### Returns: 18 | 19 | A scalar `Tensor` of type `string`. The serialized `Summary` protocol 20 | buffer. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.shape.md: -------------------------------------------------------------------------------- 1 | ### `tf.shape(input, name=None)` {#shape} 2 | 3 | Returns the shape of a tensor. 4 | 5 | This operation returns a 1-D integer tensor representing the shape of `input`. 6 | 7 | For example: 8 | 9 | ```prettyprint 10 | # 't' is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]] 11 | shape(t) ==> [2, 2, 3] 12 | ``` 13 | 14 | ##### Args: 15 | 16 | 17 | * `input`: A `Tensor`. 18 | * `name`: A name for the operation (optional). 19 | 20 | ##### Returns: 21 | 22 | A `Tensor` of type `int32`. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.to_double.md: -------------------------------------------------------------------------------- 1 | ### `tf.to_double(x, name='ToDouble')` {#to_double} 2 | 3 | Casts a tensor to type `float64`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` or `SparseTensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` or `SparseTensor` with same shape as `x` with type `float64`. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `TypeError`: If `x` cannot be cast to the `float64`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard3/tf.trace.md: -------------------------------------------------------------------------------- 1 | ### `tf.trace(x, name=None)` {#trace} 2 | 3 | Compute the trace of a tensor `x`. 4 | 5 | `trace(x)` returns the sum of along the diagonal. 6 | 7 | For example: 8 | 9 | ```python 10 | # 'x' is [[1, 1], 11 | # [1, 1]] 12 | tf.trace(x) ==> 2 13 | 14 | # 'x' is [[1,2,3], 15 | # [4,5,6], 16 | # [7,8,9]] 17 | tf.trace(x) ==> 15 18 | ``` 19 | 20 | ##### Args: 21 | 22 | 23 | * `x`: 2-D tensor. 24 | * `name`: A name for the operation (optional). 25 | 26 | ##### Returns: 27 | 28 | The trace of input tensor. 29 | 30 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.argmin.md: -------------------------------------------------------------------------------- 1 | ### `tf.argmin(input, dimension, name=None)` {#argmin} 2 | 3 | Returns the index with the smallest value across dimensions of a tensor. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 9 | * `dimension`: A `Tensor` of type `int32`. 10 | int32, 0 <= dimension < rank(input). Describes which dimension 11 | of the input Tensor to reduce across. For vectors, use dimension = 0. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A `Tensor` of type `int64`. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.batch_ifft.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_ifft(input, name=None)` {#batch_ifft} 2 | 3 | Compute the inverse 1-dimensional discrete Fourier Transform over the inner-most 4 | 5 | dimension of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 17 | dimension of `input` is replaced with its inverse 1D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.clip_by_value.md: -------------------------------------------------------------------------------- 1 | ### `tf.clip_by_value(t, clip_value_min, clip_value_max, name=None)` {#clip_by_value} 2 | 3 | Clips tensor values to a specified min and max. 4 | 5 | Given a tensor `t`, this operation returns a tensor of the same type and 6 | shape as `t` with its values clipped to `clip_value_min` and `clip_value_max`. 7 | Any values less than `clip_value_min` are set to `clip_value_min`. Any values 8 | greater than `clip_value_max` are set to `clip_value_max`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `t`: A `Tensor`. 14 | * `clip_value_min`: A 0-D (scalar) `Tensor`. The minimum value to clip by. 15 | * `clip_value_max`: A 0-D (scalar) `Tensor`. The maximum value to clip by. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A clipped `Tensor`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.framework.add_arg_scope.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.add_arg_scope(func)` {#add_arg_scope} 2 | 3 | Decorates a function with args so it can be used within an arg_scope. 4 | 5 | ##### Args: 6 | 7 | 8 | * `func`: function to decorate. 9 | 10 | ##### Returns: 11 | 12 | A tuple with the decorated function func_with_args(). 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.contrib.util.stripped_op_list_for_graph.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.util.stripped_op_list_for_graph(graph_def)` {#stripped_op_list_for_graph} 2 | 3 | Collect the stripped OpDefs for ops used by a graph. 4 | 5 | This function computes the `stripped_op_list` field of `MetaGraphDef` and 6 | similar protos. The result can be communicated from the producer to the 7 | consumer, which can then use the C++ function 8 | `RemoveNewDefaultAttrsFromGraphDef` to improve forwards compatibility. 9 | 10 | ##### Args: 11 | 12 | 13 | * `graph_def`: A `GraphDef` proto, as from `graph.as_graph_def()`. 14 | 15 | ##### Returns: 16 | 17 | An `OpList` of ops used by the graph. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `ValueError`: If an unregistered op is used. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.errors.FailedPreconditionError.md: -------------------------------------------------------------------------------- 1 | Operation was rejected because the system is not in a state to execute it. 2 | 3 | This exception is most commonly raised when running an operation 4 | that reads a [`tf.Variable`](../../api_docs/python/state_ops.md#Variable) 5 | before it has been initialized. 6 | 7 | - - - 8 | 9 | #### `tf.errors.FailedPreconditionError.__init__(node_def, op, message)` {#FailedPreconditionError.__init__} 10 | 11 | Creates a `FailedPreconditionError`. 12 | 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.floor.md: -------------------------------------------------------------------------------- 1 | ### `tf.floor(x, name=None)` {#floor} 2 | 3 | Returns element-wise largest integer not greater than x. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.greater.md: -------------------------------------------------------------------------------- 1 | ### `tf.greater(x, y, name=None)` {#greater} 2 | 3 | Returns the truth value of (x > y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.image.hsv_to_rgb.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.hsv_to_rgb(images, name=None)` {#hsv_to_rgb} 2 | 3 | Convert one or more images from HSV to RGB. 4 | 5 | Outputs a tensor of the same shape as the `images` tensor, containing the RGB 6 | value of the pixels. The output is only well defined if the value in `images` 7 | are in `[0,1]`. 8 | 9 | See `rgb_to_hsv` for a description of the HSV encoding. 10 | 11 | ##### Args: 12 | 13 | 14 | * `images`: A `Tensor` of type `float32`. 15 | 1-D or higher rank. HSV data to convert. Last dimension must be size 3. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor` of type `float32`. `images` converted to RGB. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.initialize_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.initialize_variables(var_list, name='init')` {#initialize_variables} 2 | 3 | Returns an Op that initializes a list of variables. 4 | 5 | After you launch the graph in a session, you can run the returned Op to 6 | initialize all the variables in `var_list`. This Op runs all the 7 | initializers of the variables in `var_list` in parallel. 8 | 9 | Calling `initialize_variables()` is equivalent to passing the list of 10 | initializers to `Group()`. 11 | 12 | If `var_list` is empty, however, the function still returns an Op that can 13 | be run. That Op just has no effect. 14 | 15 | ##### Args: 16 | 17 | 18 | * `var_list`: List of `Variable` objects to initialize. 19 | * `name`: Optional name for the returned operation. 20 | 21 | ##### Returns: 22 | 23 | An Op that run the initializers of all the specified variables. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.log.md: -------------------------------------------------------------------------------- 1 | ### `tf.log(x, name=None)` {#log} 2 | 3 | Computes natural logarithm of x element-wise. 4 | 5 | I.e., \\(y = \log_e x\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.l2_loss.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.l2_loss(t, name=None)` {#l2_loss} 2 | 3 | L2 Loss. 4 | 5 | Computes half the L2 norm of a tensor without the `sqrt`: 6 | 7 | output = sum(t ** 2) / 2 8 | 9 | ##### Args: 10 | 11 | 12 | * `t`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 13 | Typically 2-D, but may have any dimensions. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | A `Tensor`. Has the same type as `t`. 0-D. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.nn.softplus.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.softplus(features, name=None)` {#softplus} 2 | 3 | Computes softplus: `log(exp(features) + 1)`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `features`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `features`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.placeholder_with_default.md: -------------------------------------------------------------------------------- 1 | ### `tf.placeholder_with_default(input, shape, name=None)` {#placeholder_with_default} 2 | 3 | A placeholder op that passes though `input` when its output is not fed. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor`. The default value to produce when `output` is not fed. 9 | * `shape`: A `tf.TensorShape` or list of `ints`. 10 | The (possibly partial) shape of the tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `input`. 16 | A placeholder tensor that defaults to `input` if it is not fed. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.sub.md: -------------------------------------------------------------------------------- 1 | ### `tf.sub(x, y, name=None)` {#sub} 2 | 3 | Returns x - y element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.tile.md: -------------------------------------------------------------------------------- 1 | ### `tf.tile(input, multiples, name=None)` {#tile} 2 | 3 | Constructs a tensor by tiling a given tensor. 4 | 5 | This operation creates a new tensor by replicating `input` `multiples` times. 6 | The output tensor's i'th dimension has `input.dims(i) * multiples[i]` elements, 7 | and the values of `input` are replicated `multiples[i]` times along the 'i'th 8 | dimension. For example, tiling `[a b c d]` by `[2]` produces 9 | `[a b c d a b c d]`. 10 | 11 | ##### Args: 12 | 13 | 14 | * `input`: A `Tensor`. 1-D or higher. 15 | * `multiples`: A `Tensor` of type `int32`. 16 | 1-D. Length must be the same as the number of dimensions in `input` 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A `Tensor`. Has the same type as `input`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.train.match_filenames_once.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.match_filenames_once(pattern, name=None)` {#match_filenames_once} 2 | 3 | Save the list of files matching pattern, so it is only computed once. 4 | 5 | ##### Args: 6 | 7 | 8 | * `pattern`: A file pattern (glob). 9 | * `name`: A name for the operations (optional). 10 | 11 | ##### Returns: 12 | 13 | A variable that is initialized to the list of files matching pattern. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard4/tf.train.write_graph.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.write_graph(graph_def, logdir, name, as_text=True)` {#write_graph} 2 | 3 | Writes a graph proto to a file. 4 | 5 | The graph is written as a binary proto unless `as_text` is `True`. 6 | 7 | ```python 8 | v = tf.Variable(0, name='my_variable') 9 | sess = tf.Session() 10 | tf.train.write_graph(sess.graph_def, '/tmp/my-model', 'train.pbtxt') 11 | ``` 12 | 13 | ##### Args: 14 | 15 | 16 | * `graph_def`: A `GraphDef` protocol buffer. 17 | * `logdir`: Directory where to write the graph. This can refer to remote 18 | filesystems, such as Google Cloud Storage (GCS). 19 | * `name`: Filename for the graph. 20 | * `as_text`: If `True`, writes the graph as an ASCII proto. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.add.md: -------------------------------------------------------------------------------- 1 | ### `tf.add(x, y, name=None)` {#add} 2 | 3 | Returns x + y element-wise. 4 | 5 | *NOTE*: Add supports broadcasting. AddN does not. 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`, `string`. 11 | * `y`: A `Tensor`. Must have the same type as `x`. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A `Tensor`. Has the same type as `x`. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.asin.md: -------------------------------------------------------------------------------- 1 | ### `tf.asin(x, name=None)` {#asin} 2 | 3 | Computes asin of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.bytes.md: -------------------------------------------------------------------------------- 1 | str(object='') -> string 2 | 3 | Return a nice string representation of the object. 4 | If the argument is a string, the return value is the same object. 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.cast.md: -------------------------------------------------------------------------------- 1 | ### `tf.cast(x, dtype, name=None)` {#cast} 2 | 3 | Casts a tensor to a new type. 4 | 5 | The operation casts `x` (in case of `Tensor`) or `x.values` 6 | (in case of `SparseTensor`) to `dtype`. 7 | 8 | For example: 9 | 10 | ```python 11 | # tensor `a` is [1.8, 2.2], dtype=tf.float 12 | tf.cast(a, tf.int32) ==> [1, 2] # dtype=tf.int32 13 | ``` 14 | 15 | ##### Args: 16 | 17 | 18 | * `x`: A `Tensor` or `SparseTensor`. 19 | * `dtype`: The destination type. 20 | * `name`: A name for the operation (optional). 21 | 22 | ##### Returns: 23 | 24 | A `Tensor` or `SparseTensor` with same shape as `x`. 25 | 26 | ##### Raises: 27 | 28 | 29 | * `TypeError`: If `x` cannot be cast to the `dtype`. 30 | 31 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.framework.arg_scoped_arguments.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.arg_scoped_arguments(func)` {#arg_scoped_arguments} 2 | 3 | Returns the list kwargs that arg_scope can set for a func. 4 | 5 | ##### Args: 6 | 7 | 8 | * `func`: function which has been decorated with @add_arg_scope. 9 | 10 | ##### Returns: 11 | 12 | a list of kwargs names. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.framework.get_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_variables(scope=None, suffix=None, collection='variables')` {#get_variables} 2 | 3 | Gets the list of variables, filtered by scope and/or suffix. 4 | 5 | ##### Args: 6 | 7 | 8 | * `scope`: an optional scope for filtering the variables to return. 9 | * `suffix`: an optional suffix for filtering the variables to return. 10 | * `collection`: in which collection search for. Defaults to GraphKeys.VARIABLES. 11 | 12 | ##### Returns: 13 | 14 | a list of variables in colelction with scope and suffix. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.layers.sum_regularizer.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.sum_regularizer(regularizer_list)` {#sum_regularizer} 2 | 3 | Returns a function that applies the sum of multiple regularizers. 4 | 5 | ##### Args: 6 | 7 | 8 | * `regularizer_list`: A list of regularizers to apply. 9 | 10 | ##### Returns: 11 | 12 | A function with signature `sum_reg(weights, name=None)` that applies the 13 | sum of all the input regularizers. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.layers.summarize_collection.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.summarize_collection(collection, name_filter=None, summarizer=summarize_tensor)` {#summarize_collection} 2 | 3 | Summarize a graph collection of tensors, possibly filtered by name. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.learn.run_n.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.run_n(output_dict, feed_dict=None, restore_checkpoint_path=None, n=1)` {#run_n} 2 | 3 | Run `output_dict` tensors `n` times, with the same `feed_dict` each run. 4 | 5 | ##### Args: 6 | 7 | 8 | * `output_dict`: A `dict` mapping string names to tensors to run. Must all be 9 | from the same graph. 10 | * `feed_dict`: `dict` of input values to feed each run. 11 | * `restore_checkpoint_path`: A string containing the path to a checkpoint to 12 | restore. 13 | * `n`: Number of times to repeat. 14 | 15 | ##### Returns: 16 | 17 | A list of `n` `dict` objects, each containing values read from `output_dict` 18 | tensors. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.losses.sigmoid_cross_entropy.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.losses.sigmoid_cross_entropy(logits, multi_class_labels, weight=1.0, label_smoothing=0, scope=None)` {#sigmoid_cross_entropy} 2 | 3 | Creates a cross-entropy loss using tf.nn.sigmoid_cross_entropy_with_logits. 4 | 5 | ##### Args: 6 | 7 | 8 | * `logits`: [batch_size, num_classes] logits outputs of the network . 9 | * `multi_class_labels`: [batch_size, num_classes] target labels in (0, 1). 10 | * `weight`: Coefficients for the loss. The tensor must be a scalar, a tensor of 11 | shape [batch_size] or shape [batch_size, num_classes]. 12 | * `label_smoothing`: If greater than 0 then smooth the labels. 13 | * `scope`: The scope for the operations performed in computing the loss. 14 | 15 | ##### Returns: 16 | 17 | A scalar `Tensor` representing the loss value. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.contrib.metrics.accuracy.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.metrics.accuracy(predictions, labels, weights=None)` {#accuracy} 2 | 3 | Computes the percentage of times that predictions matches labels. 4 | 5 | ##### Args: 6 | 7 | 8 | * `predictions`: the predicted values, a `Tensor` whose dtype and shape 9 | matches 'labels'. 10 | * `labels`: the ground truth values, a `Tensor` of any shape and 11 | integer or string dtype. 12 | * `weights`: None or `Tensor` of float values to reweight the accuracy. 13 | 14 | ##### Returns: 15 | 16 | Accuracy `Tensor`. 17 | 18 | ##### Raises: 19 | 20 | 21 | * `ValueError`: if dtypes don't match or 22 | if dtype is not integer or string. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.cos.md: -------------------------------------------------------------------------------- 1 | ### `tf.cos(x, name=None)` {#cos} 2 | 3 | Computes cos of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.div.md: -------------------------------------------------------------------------------- 1 | ### `tf.div(x, y, name=None)` {#div} 2 | 3 | Returns x / y element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.errors.PermissionDeniedError.md: -------------------------------------------------------------------------------- 1 | Raised when the caller does not have permission to run an operation. 2 | 3 | For example, running the 4 | [`tf.WholeFileReader.read()`](../../api_docs/python/io_ops.md#WholeFileReader) 5 | operation could raise `PermissionDeniedError` if it receives the name of a 6 | file for which the user does not have the read file permission. 7 | 8 | - - - 9 | 10 | #### `tf.errors.PermissionDeniedError.__init__(node_def, op, message)` {#PermissionDeniedError.__init__} 11 | 12 | Creates a `PermissionDeniedError`. 13 | 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.errors.UnavailableError.md: -------------------------------------------------------------------------------- 1 | Raised when the runtime is currently unavailable. 2 | 3 | This exception is not currently used. 4 | 5 | - - - 6 | 7 | #### `tf.errors.UnavailableError.__init__(node_def, op, message)` {#UnavailableError.__init__} 8 | 9 | Creates an `UnavailableError`. 10 | 11 | 12 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.image.flip_left_right.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.flip_left_right(image)` {#flip_left_right} 2 | 3 | Flip an image horizontally (left to right). 4 | 5 | Outputs the contents of `image` flipped along the second dimension, which is 6 | `width`. 7 | 8 | See also `reverse()`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `image`: A 3-D tensor of shape `[height, width, channels].` 14 | 15 | ##### Returns: 16 | 17 | A 3-D tensor of the same type and shape as `image`. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `ValueError`: if the shape of `image` not supported. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.image.flip_up_down.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.flip_up_down(image)` {#flip_up_down} 2 | 3 | Flip an image horizontally (upside down). 4 | 5 | Outputs the contents of `image` flipped along the first dimension, which is 6 | `height`. 7 | 8 | See also `reverse()`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `image`: A 3-D tensor of shape `[height, width, channels].` 14 | 15 | ##### Returns: 16 | 17 | A 3-D tensor of the same type and shape as `image`. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `ValueError`: if the shape of `image` not supported. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.inv.md: -------------------------------------------------------------------------------- 1 | ### `tf.inv(x, name=None)` {#inv} 2 | 3 | Computes the reciprocal of x element-wise. 4 | 5 | I.e., \\(y = 1 / x\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.load_file_system_library.md: -------------------------------------------------------------------------------- 1 | ### `tf.load_file_system_library(library_filename)` {#load_file_system_library} 2 | 3 | Loads a TensorFlow plugin, containing file system implementation. 4 | 5 | Pass `library_filename` to a platform-specific mechanism for dynamically 6 | loading a library. The rules for determining the exact location of the 7 | library are platform-specific and are not documented here. 8 | 9 | ##### Args: 10 | 11 | 12 | * `library_filename`: Path to the plugin. 13 | Relative or absolute filesystem path to a dynamic library file. 14 | 15 | ##### Returns: 16 | 17 | None. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `RuntimeError`: when unable to load the library. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_and.md: -------------------------------------------------------------------------------- 1 | ### `tf.logical_and(x, y, name=None)` {#logical_and} 2 | 3 | Returns the truth value of x AND y element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` of type `bool`. 9 | * `y`: A `Tensor` of type `bool`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.logical_not.md: -------------------------------------------------------------------------------- 1 | ### `tf.logical_not(x, name=None)` {#logical_not} 2 | 3 | Returns the truth value of NOT x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` of type `bool`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `bool`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.no_op.md: -------------------------------------------------------------------------------- 1 | ### `tf.no_op(name=None)` {#no_op} 2 | 3 | Does nothing. Only useful as a placeholder for control edges. 4 | 5 | ##### Args: 6 | 7 | 8 | * `name`: A name for the operation (optional). 9 | 10 | ##### Returns: 11 | 12 | The created Operation. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.sign.md: -------------------------------------------------------------------------------- 1 | ### `tf.sign(x, name=None)` {#sign} 2 | 3 | Returns an element-wise indication of the sign of a number. 4 | 5 | `y = sign(x) = -1` if `x < 0`; 0 if `x == 0`; 1 if `x > 0`. 6 | 7 | For complex numbers, `y = sign(x) = x / |x|` if `x != 0`, otherwise `y = 0`. 8 | 9 | ##### Args: 10 | 11 | 12 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 13 | * `name`: A name for the operation (optional). 14 | 15 | ##### Returns: 16 | 17 | A `Tensor`. Has the same type as `x`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.tan.md: -------------------------------------------------------------------------------- 1 | ### `tf.tan(x, name=None)` {#tan} 2 | 3 | Computes tan of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.test.main.md: -------------------------------------------------------------------------------- 1 | ### `tf.test.main()` {#main} 2 | 3 | Runs all unit tests. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.to_int64.md: -------------------------------------------------------------------------------- 1 | ### `tf.to_int64(x, name='ToInt64')` {#to_int64} 2 | 3 | Casts a tensor to type `int64`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` or `SparseTensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` or `SparseTensor` with same shape as `x` with type `int64`. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `TypeError`: If `x` cannot be cast to the `int64`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.train.LooperThread.loop.md: -------------------------------------------------------------------------------- 1 | #### `tf.train.LooperThread.loop(coord, timer_interval_secs, target, args=None, kwargs=None)` {#LooperThread.loop} 2 | 3 | Start a LooperThread that calls a function periodically. 4 | 5 | If `timer_interval_secs` is None the thread calls `target(args)` 6 | repeatedly. Otherwise `target(args)` is called every `timer_interval_secs` 7 | seconds. The thread terminates when a stop of the coordinator is 8 | requested. 9 | 10 | ##### Args: 11 | 12 | 13 | * `coord`: A Coordinator. 14 | * `timer_interval_secs`: Number. Time boundaries at which to call `target`. 15 | * `target`: A callable object. 16 | * `args`: Optional arguments to pass to `target` when calling it. 17 | * `kwargs`: Optional keyword arguments to pass to `target` when calling it. 18 | 19 | ##### Returns: 20 | 21 | The started thread. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard5/tf.verify_tensor_all_finite.md: -------------------------------------------------------------------------------- 1 | ### `tf.verify_tensor_all_finite(t, msg, name=None)` {#verify_tensor_all_finite} 2 | 3 | Assert that the tensor does not contain any NaN's or Inf's. 4 | 5 | ##### Args: 6 | 7 | 8 | * `t`: Tensor to check. 9 | * `msg`: Message to log on failure. 10 | * `name`: A name for this operation (optional). 11 | 12 | ##### Returns: 13 | 14 | Same tensor as `t`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.FixedLenSequenceFeature.md: -------------------------------------------------------------------------------- 1 | Configuration for a dense input feature in a sequence item. 2 | 3 | To treat a sparse input as dense, provide `allow_missing=True`; otherwise, 4 | the parse functions will fail on any examples missing this feature. 5 | 6 | Fields: 7 | shape: Shape of input data. 8 | dtype: Data type of input. 9 | allow_missing: Whether to allow this feature to be missing from a feature 10 | list item. 11 | - - - 12 | 13 | #### `tf.FixedLenSequenceFeature.allow_missing` {#FixedLenSequenceFeature.allow_missing} 14 | 15 | Alias for field number 2 16 | 17 | 18 | - - - 19 | 20 | #### `tf.FixedLenSequenceFeature.dtype` {#FixedLenSequenceFeature.dtype} 21 | 22 | Alias for field number 1 23 | 24 | 25 | - - - 26 | 27 | #### `tf.FixedLenSequenceFeature.shape` {#FixedLenSequenceFeature.shape} 28 | 29 | Alias for field number 0 30 | 31 | 32 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.abs.md: -------------------------------------------------------------------------------- 1 | ### `tf.abs(x, name=None)` {#abs} 2 | 3 | Computes the absolute value of a tensor. 4 | 5 | Given a tensor of real numbers `x`, this operation returns a tensor 6 | containing the absolute value of each element in `x`. For example, if x is 7 | an input element and y is an output element, this operation computes 8 | \\(y = |x|\\). 9 | 10 | See [`tf.complex_abs()`](#tf_complex_abs) to compute the absolute value of a complex 11 | number. 12 | 13 | ##### Args: 14 | 15 | 16 | * `x`: A `Tensor` of type `float`, `double`, `int32`, or `int64`. 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A `Tensor` the same size and type as `x` with absolute values. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.batch_fft3d.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_fft3d(input, name=None)` {#batch_fft3d} 2 | 3 | Compute the 3-dimensional discrete Fourier Transform over the inner-most 3 4 | 5 | dimensions of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 3 17 | dimensions of `input` are replaced with their 3D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.batch_ifft2d.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_ifft2d(input, name=None)` {#batch_ifft2d} 2 | 3 | Compute the inverse 2-dimensional discrete Fourier Transform over the inner-most 4 | 5 | 2 dimensions of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 2 17 | dimensions of `input` are replaced with their inverse 2D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.copy_graph.copy_variable_to_graph.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.copy_graph.copy_variable_to_graph(org_instance, to_graph, scope='')` {#copy_variable_to_graph} 2 | 3 | Given a `Variable` instance from one `Graph`, initializes and returns 4 | a copy of it from another `Graph`, under the specified scope 5 | (default `""`). 6 | 7 | Args: 8 | org_instance: A `Variable` from some `Graph`. 9 | to_graph: The `Graph` to copy the `Variable` to. 10 | scope: A scope for the new `Variable` (default `""`). 11 | 12 | ##### Returns: 13 | 14 | The copied `Variable` from `to_graph`. 15 | 16 | ##### Raises: 17 | 18 | 19 | * `TypeError`: If `org_instance` is not a `Variable`. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.framework.create_global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.create_global_step(graph=None)` {#create_global_step} 2 | 3 | Create global step tensor in graph. 4 | 5 | ##### Args: 6 | 7 | 8 | * `graph`: The graph in which to create the global step. If missing, use default 9 | graph. 10 | 11 | ##### Returns: 12 | 13 | Global step tensor. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `ValueError`: if global step key is already defined. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.framework.reduce_sum_n.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.reduce_sum_n(tensors, name=None)` {#reduce_sum_n} 2 | 3 | Reduce tensors to a scalar sum. 4 | 5 | This reduces each tensor in `tensors` to a scalar via `tf.reduce_sum`, then 6 | adds them via `tf.add_n`. 7 | 8 | ##### Args: 9 | 10 | 11 | * `tensors`: List of tensors, all of the same numeric type. 12 | * `name`: Tensor name, and scope for all other ops. 13 | 14 | ##### Returns: 15 | 16 | Total loss tensor, or None if no losses have been configured. 17 | 18 | ##### Raises: 19 | 20 | 21 | * `ValueError`: if `losses` is missing or empty. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.layers.summarize_activations.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.summarize_activations(name_filter=None, summarizer=summarize_activation)` {#summarize_activations} 2 | 3 | Summarize activations, using `summarize_activation` to summarize. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.contrib.learn.extract_dask_labels.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.extract_dask_labels(labels)` {#extract_dask_labels} 2 | 3 | Extract data from dask.Series for labels. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.erfc.md: -------------------------------------------------------------------------------- 1 | ### `tf.erfc(x, name=None)` {#erfc} 2 | 3 | Computes the complementary error function of `x` element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.errors.AbortedError.md: -------------------------------------------------------------------------------- 1 | The operation was aborted, typically due to a concurrent action. 2 | 3 | For example, running a 4 | [`queue.enqueue()`](../../api_docs/python/io_ops.md#QueueBase.enqueue) 5 | operation may raise `AbortedError` if a 6 | [`queue.close()`](../../api_docs/python/io_ops.md#QueueBase.close) operation 7 | previously ran. 8 | 9 | - - - 10 | 11 | #### `tf.errors.AbortedError.__init__(node_def, op, message)` {#AbortedError.__init__} 12 | 13 | Creates an `AbortedError`. 14 | 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.errors.InternalError.md: -------------------------------------------------------------------------------- 1 | Raised when the system experiences an internal error. 2 | 3 | This exception is raised when some invariant expected by the runtime 4 | has been broken. Catching this exception is not recommended. 5 | 6 | - - - 7 | 8 | #### `tf.errors.InternalError.__init__(node_def, op, message)` {#InternalError.__init__} 9 | 10 | Creates an `InternalError`. 11 | 12 | 13 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.errors.NotFoundError.md: -------------------------------------------------------------------------------- 1 | Raised when a requested entity (e.g., a file or directory) was not found. 2 | 3 | For example, running the 4 | [`tf.WholeFileReader.read()`](../../api_docs/python/io_ops.md#WholeFileReader) 5 | operation could raise `NotFoundError` if it receives the name of a file that 6 | does not exist. 7 | 8 | - - - 9 | 10 | #### `tf.errors.NotFoundError.__init__(node_def, op, message)` {#NotFoundError.__init__} 11 | 12 | Creates a `NotFoundError`. 13 | 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.errors.UnimplementedError.md: -------------------------------------------------------------------------------- 1 | Raised when an operation has not been implemented. 2 | 3 | Some operations may raise this error when passed otherwise-valid 4 | arguments that it does not currently support. For example, running 5 | the [`tf.nn.max_pool()`](../../api_docs/python/nn.md#max_pool) operation 6 | would raise this error if pooling was requested on the batch dimension, 7 | because this is not yet supported. 8 | 9 | - - - 10 | 11 | #### `tf.errors.UnimplementedError.__init__(node_def, op, message)` {#UnimplementedError.__init__} 12 | 13 | Creates an `UnimplementedError`. 14 | 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.igamma.md: -------------------------------------------------------------------------------- 1 | ### `tf.igamma(a, x, name=None)` {#igamma} 2 | 3 | Compute the lower regularized incomplete Gamma function `Q(a, x)`. 4 | 5 | The lower regularized incomplete Gamma function is defined as: 6 | 7 | ``` 8 | P(a, x) = gamma(a, x) / Gamma(x) = 1 - Q(a, x) 9 | ``` 10 | where 11 | ``` 12 | gamma(a, x) = int_{0}^{x} t^{a-1} exp(-t) dt 13 | ``` 14 | is the lower incomplete Gamma function. 15 | 16 | Note, above `Q(a, x)` (`Igammac`) is the upper regularized complete 17 | Gamma function. 18 | 19 | ##### Args: 20 | 21 | 22 | * `a`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 23 | * `x`: A `Tensor`. Must have the same type as `a`. 24 | * `name`: A name for the operation (optional). 25 | 26 | ##### Returns: 27 | 28 | A `Tensor`. Has the same type as `a`. 29 | 30 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.image.rgb_to_hsv.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.rgb_to_hsv(images, name=None)` {#rgb_to_hsv} 2 | 3 | Converts one or more images from RGB to HSV. 4 | 5 | Outputs a tensor of the same shape as the `images` tensor, containing the HSV 6 | value of the pixels. The output is only well defined if the value in `images` 7 | are in `[0,1]`. 8 | 9 | `output[..., 0]` contains hue, `output[..., 1]` contains saturation, and 10 | `output[..., 2]` contains value. All HSV values are in `[0,1]`. A hue of 0 11 | corresponds to pure red, hue 1/3 is pure green, and 2/3 is pure blue. 12 | 13 | ##### Args: 14 | 15 | 16 | * `images`: A `Tensor` of type `float32`. 17 | 1-D or higher rank. RGB data to convert. Last dimension must be size 3. 18 | * `name`: A name for the operation (optional). 19 | 20 | ##### Returns: 21 | 22 | A `Tensor` of type `float32`. `images` converted to HSV. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.load_op_library.md: -------------------------------------------------------------------------------- 1 | ### `tf.load_op_library(library_filename)` {#load_op_library} 2 | 3 | Loads a TensorFlow plugin, containing custom ops and kernels. 4 | 5 | Pass "library_filename" to a platform-specific mechanism for dynamically 6 | loading a library. The rules for determining the exact location of the 7 | library are platform-specific and are not documented here. 8 | 9 | ##### Args: 10 | 11 | 12 | * `library_filename`: Path to the plugin. 13 | Relative or absolute filesystem path to a dynamic library file. 14 | 15 | ##### Returns: 16 | 17 | A python module containing the Python wrappers for Ops defined in 18 | the plugin. 19 | 20 | ##### Raises: 21 | 22 | 23 | * `RuntimeError`: when unable to load the library or get the python wrappers. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.maximum.md: -------------------------------------------------------------------------------- 1 | ### `tf.maximum(x, y, name=None)` {#maximum} 2 | 3 | Returns the max of x and y (i.e. x > y ? x : y) element-wise, broadcasts. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.moving_average_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.moving_average_variables()` {#moving_average_variables} 2 | 3 | Returns all variables that maintain their moving averages. 4 | 5 | If an `ExponentialMovingAverage` object is created and the `apply()` 6 | method is called on a list of variables, these variables will 7 | be added to the `GraphKeys.MOVING_AVERAGE_VARIABLES` collection. 8 | This convenience function returns the contents of that collection. 9 | 10 | ##### Returns: 11 | 12 | A list of Variable objects. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.elu.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.elu(features, name=None)` {#elu} 2 | 3 | Computes exponential linear: `exp(features) - 1` if < 0, `features` otherwise. 4 | 5 | See [Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs) 6 | ](http://arxiv.org/abs/1511.07289) 7 | 8 | ##### Args: 9 | 10 | 11 | * `features`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A `Tensor`. Has the same type as `features`. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.nn.softmax.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.softmax(logits, name=None)` {#softmax} 2 | 3 | Computes softmax activations. 4 | 5 | For each batch `i` and class `j` we have 6 | 7 | softmax[i, j] = exp(logits[i, j]) / sum(exp(logits[i])) 8 | 9 | ##### Args: 10 | 11 | 12 | * `logits`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 13 | 2-D with shape `[batch_size, num_classes]`. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | A `Tensor`. Has the same type as `logits`. Same shape as `logits`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.pow.md: -------------------------------------------------------------------------------- 1 | ### `tf.pow(x, y, name=None)` {#pow} 2 | 3 | Computes the power of one value to another. 4 | 5 | Given a tensor `x` and a tensor `y`, this operation computes \\(x^y\\) for 6 | corresponding elements in `x` and `y`. For example: 7 | 8 | ``` 9 | # tensor 'x' is [[2, 2], [3, 3]] 10 | # tensor 'y' is [[8, 16], [2, 3]] 11 | tf.pow(x, y) ==> [[256, 65536], [9, 27]] 12 | ``` 13 | 14 | ##### Args: 15 | 16 | 17 | * `x`: A `Tensor` of type `float`, `double`, `int32`, `int64`, `complex64`, or 18 | `complex128`. 19 | * `y`: A `Tensor` of type `float`, `double`, `int32`, `int64`, `complex64`, or 20 | `complex128`. 21 | * `name`: A name for the operation (optional). 22 | 23 | ##### Returns: 24 | 25 | A `Tensor`. 26 | 27 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.python_io.tf_record_iterator.md: -------------------------------------------------------------------------------- 1 | ### `tf.python_io.tf_record_iterator(path)` {#tf_record_iterator} 2 | 3 | An iterator that read the records from a TFRecords file. 4 | 5 | ##### Args: 6 | 7 | 8 | * `path`: The path to the TFRecords file. 9 | 10 | ##### Yields: 11 | 12 | Strings. 13 | 14 | ##### Raises: 15 | 16 | 17 | * `IOError`: If `path` cannot be opened for reading. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.rank.md: -------------------------------------------------------------------------------- 1 | ### `tf.rank(input, name=None)` {#rank} 2 | 3 | Returns the rank of a tensor. 4 | 5 | This operation returns an integer representing the rank of `input`. 6 | 7 | For example: 8 | 9 | ```python 10 | # 't' is [[[1, 1, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]] 11 | # shape of tensor 't' is [2, 2, 3] 12 | rank(t) ==> 3 13 | ``` 14 | 15 | **Note**: The rank of a tensor is not the same as the rank of a matrix. The 16 | rank of a tensor is the number of indices required to uniquely select each 17 | element of the tensor. Rank is also known as "order", "degree", or "ndims." 18 | 19 | ##### Args: 20 | 21 | 22 | * `input`: A `Tensor` or `SparseTensor`. 23 | * `name`: A name for the operation (optional). 24 | 25 | ##### Returns: 26 | 27 | A `Tensor` of type `int32`. 28 | 29 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.self_adjoint_eig.md: -------------------------------------------------------------------------------- 1 | ### `tf.self_adjoint_eig(input, name=None)` {#self_adjoint_eig} 2 | 3 | Calculates the Eigen Decomposition of a square Self-Adjoint matrix. 4 | 5 | Only the lower-triangular part of the input will be used in this case. The 6 | upper-triangular part will not be read. 7 | 8 | The result is a M+1 x M matrix whose first row is the eigenvalues, and 9 | subsequent rows are eigenvectors. 10 | 11 | ##### Args: 12 | 13 | 14 | * `input`: A `Tensor`. Must be one of the following types: `float64`, `float32`. 15 | Shape is `[M, M]`. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor`. Has the same type as `input`. Shape is `[M+1, M]`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.sigmoid.md: -------------------------------------------------------------------------------- 1 | ### `tf.sigmoid(x, name=None)` {#sigmoid} 2 | 3 | Computes sigmoid of `x` element-wise. 4 | 5 | Specifically, `y = 1 / (1 + exp(-x))`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A Tensor with type `float`, `double`, `int32`, `complex64`, `int64`, 11 | or `qint32`. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A Tensor with the same type as `x` if `x.dtype != qint32` 17 | otherwise the return type is `quint8`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.to_float.md: -------------------------------------------------------------------------------- 1 | ### `tf.to_float(x, name='ToFloat')` {#to_float} 2 | 3 | Casts a tensor to type `float32`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` or `SparseTensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` or `SparseTensor` with same shape as `x` with type `float32`. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `TypeError`: If `x` cannot be cast to the `float32`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.train.QueueRunner.from_proto.md: -------------------------------------------------------------------------------- 1 | #### `tf.train.QueueRunner.from_proto(queue_runner_def)` {#QueueRunner.from_proto} 2 | 3 | Returns a `QueueRunner` object created from `queue_runner_def`. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard6/tf.trainable_variables.md: -------------------------------------------------------------------------------- 1 | ### `tf.trainable_variables()` {#trainable_variables} 2 | 3 | Returns all variables created with `trainable=True`. 4 | 5 | When passed `trainable=True`, the `Variable()` constructor automatically 6 | adds new variables to the graph collection 7 | `GraphKeys.TRAINABLE_VARIABLES`. This convenience function returns the 8 | contents of that collection. 9 | 10 | ##### Returns: 11 | 12 | A list of Variable objects. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.DeviceSpec.from_string.md: -------------------------------------------------------------------------------- 1 | #### `tf.DeviceSpec.from_string(spec)` {#DeviceSpec.from_string} 2 | 3 | Construct a `DeviceSpec` from a string. 4 | 5 | ##### Args: 6 | 7 | 8 | * `spec`: a string of the form 9 | /job:/replica:/task:/device:CPU: 10 | or 11 | /job:/replica:/task:/device:GPU: 12 | as cpu and gpu are mutually exclusive. 13 | All entries are optional. 14 | 15 | ##### Returns: 16 | 17 | A DeviceSpec. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.FixedLenFeature.md: -------------------------------------------------------------------------------- 1 | Configuration for parsing a fixed-length input feature. 2 | 3 | To treat sparse input as dense, provide a `default_value`; otherwise, 4 | the parse functions will fail on any examples missing this feature. 5 | 6 | Fields: 7 | shape: Shape of input data. 8 | dtype: Data type of input. 9 | default_value: Value to be used if an example is missing this feature. It 10 | must be compatible with `dtype`. 11 | - - - 12 | 13 | #### `tf.FixedLenFeature.default_value` {#FixedLenFeature.default_value} 14 | 15 | Alias for field number 2 16 | 17 | 18 | - - - 19 | 20 | #### `tf.FixedLenFeature.dtype` {#FixedLenFeature.dtype} 21 | 22 | Alias for field number 1 23 | 24 | 25 | - - - 26 | 27 | #### `tf.FixedLenFeature.shape` {#FixedLenFeature.shape} 28 | 29 | Alias for field number 0 30 | 31 | 32 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.QueueBase.from_list.md: -------------------------------------------------------------------------------- 1 | #### `tf.QueueBase.from_list(index, queues)` {#QueueBase.from_list} 2 | 3 | Create a queue using the queue reference from `queues[index]`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `index`: An integer scalar tensor that determines the input that gets 9 | selected. 10 | * `queues`: A list of `QueueBase` objects. 11 | 12 | ##### Returns: 13 | 14 | A `QueueBase` object. 15 | 16 | ##### Raises: 17 | 18 | 19 | * `TypeError`: When `queues` is not a list of `QueueBase` objects, 20 | or when the data types of `queues` are not all the same. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.as_dtype.md: -------------------------------------------------------------------------------- 1 | ### `tf.as_dtype(type_value)` {#as_dtype} 2 | 3 | Converts the given `type_value` to a `DType`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `type_value`: A value that can be converted to a `tf.DType` 9 | object. This may currently be a `tf.DType` object, a 10 | [`DataType` enum](https://www.tensorflow.org/code/tensorflow/core/framework/types.proto), 11 | a string type name, or a `numpy.dtype`. 12 | 13 | ##### Returns: 14 | 15 | A `DType` corresponding to `type_value`. 16 | 17 | ##### Raises: 18 | 19 | 20 | * `TypeError`: If `type_value` cannot be converted to a `DType`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.batch_cholesky.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_cholesky(input, name=None)` {#batch_cholesky} 2 | 3 | Calculates the Cholesky decomposition of a batch of square matrices. 4 | 5 | The input is a tensor of shape `[..., M, M]` whose inner-most 2 dimensions 6 | form square matrices, with the same constraints as the single matrix Cholesky 7 | decomposition above. The output is a tensor of the same shape as the input 8 | containing the Cholesky decompositions for all input submatrices `[..., :, :]`. 9 | 10 | ##### Args: 11 | 12 | 13 | * `input`: A `Tensor`. Must be one of the following types: `float64`, `float32`. 14 | Shape is `[..., M, M]`. 15 | * `name`: A name for the operation (optional). 16 | 17 | ##### Returns: 18 | 19 | A `Tensor`. Has the same type as `input`. Shape is `[..., M, M]`. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.contrib.framework.assert_or_get_global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.assert_or_get_global_step(graph=None, global_step_tensor=None)` {#assert_or_get_global_step} 2 | 3 | Verifies that a global step tensor is valid or gets one if None is given. 4 | 5 | If `global_step_tensor` is not None, check that it is a valid global step 6 | tensor (using `assert_global_step`). Otherwise find a global step tensor using 7 | `get_global_step` and return it. 8 | 9 | ##### Args: 10 | 11 | 12 | * `graph`: The graph to find the global step tensor for. 13 | * `global_step_tensor`: The tensor to check for suitability as a global step. 14 | If None is given (the default), find a global step tensor. 15 | 16 | ##### Returns: 17 | 18 | A tensor suitable as a global step, or `None` if none was provided and none 19 | was found. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.contrib.framework.get_variables_by_suffix.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_variables_by_suffix(suffix, scope=None)` {#get_variables_by_suffix} 2 | 3 | Gets the list of variables that end with the given suffix. 4 | 5 | ##### Args: 6 | 7 | 8 | * `suffix`: suffix for filtering the variables to return. 9 | * `scope`: an optional scope for filtering the variables to return. 10 | 11 | ##### Returns: 12 | 13 | a copied list of variables with the given name and prefix. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.contrib.framework.has_arg_scope.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.has_arg_scope(func)` {#has_arg_scope} 2 | 3 | Checks whether a func has been decorated with @add_arg_scope or not. 4 | 5 | ##### Args: 6 | 7 | 8 | * `func`: function to check. 9 | 10 | ##### Returns: 11 | 12 | a boolean. 13 | 14 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.contrib.framework.with_shape.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.with_shape(expected_shape, tensor)` {#with_shape} 2 | 3 | Asserts tensor has expected shape. 4 | 5 | If tensor shape and expected_shape, are fully defined, assert they match. 6 | Otherwise, add assert op that will validate the shape when tensor is 7 | evaluated, and set shape on tensor. 8 | 9 | ##### Args: 10 | 11 | 12 | * `expected_shape`: Expected shape to assert, as a 1D array of ints, or tensor 13 | of same. 14 | * `tensor`: Tensor whose shape we're validating. 15 | 16 | ##### Returns: 17 | 18 | tensor, perhaps with a dependent assert operation. 19 | 20 | ##### Raises: 21 | 22 | 23 | * `ValueError`: if tensor has an invalid shape. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.contrib.util.make_ndarray.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.util.make_ndarray(tensor)` {#make_ndarray} 2 | 3 | Create a numpy ndarray from a tensor. 4 | 5 | Create a numpy ndarray with the same shape and data as the tensor. 6 | 7 | ##### Args: 8 | 9 | 10 | * `tensor`: A TensorProto. 11 | 12 | ##### Returns: 13 | 14 | A numpy array with the tensor contents. 15 | 16 | ##### Raises: 17 | 18 | 19 | * `TypeError`: if tensor has unsupported type. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.count_up_to.md: -------------------------------------------------------------------------------- 1 | ### `tf.count_up_to(ref, limit, name=None)` {#count_up_to} 2 | 3 | Increments 'ref' until it reaches 'limit'. 4 | 5 | This operation outputs "ref" after the update is done. This makes it 6 | easier to chain operations that need to use the updated value. 7 | 8 | ##### Args: 9 | 10 | 11 | * `ref`: A mutable `Tensor`. Must be one of the following types: `int32`, `int64`. 12 | Should be from a scalar `Variable` node. 13 | * `limit`: An `int`. 14 | If incrementing ref would bring it above limit, instead generates an 15 | 'OutOfRange' error. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor`. Has the same type as `ref`. 21 | A copy of the input before increment. If nothing else modifies the 22 | input, the values produced will all be distinct. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.fft3d.md: -------------------------------------------------------------------------------- 1 | ### `tf.fft3d(input, name=None)` {#fft3d} 2 | 3 | Compute the 3-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 3-D tensor. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. The 3D Fourier Transform of `input`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.get_collection_ref.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_collection_ref(key)` {#get_collection_ref} 2 | 3 | Wrapper for `Graph.get_collection_ref()` using the default graph. 4 | 5 | See [`Graph.get_collection_ref()`](../../api_docs/python/framework.md#Graph.get_collection_ref) 6 | for more details. 7 | 8 | ##### Args: 9 | 10 | 11 | * `key`: The key for the collection. For example, the `GraphKeys` class 12 | contains many standard names for collections. 13 | 14 | ##### Returns: 15 | 16 | The list of values in the collection with the given `name`, or an empty 17 | list if no value has been added to that collection. Note that this returns 18 | the collection list itself, which can be modified in place to change the 19 | collection. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.get_session_tensor.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_session_tensor(dtype, name=None)` {#get_session_tensor} 2 | 3 | Get the tensor of type `dtype` by feeding a tensor handle. 4 | 5 | This is EXPERIMENTAL and subject to change. 6 | 7 | Get the value of the tensor from a tensor handle. The tensor 8 | is produced in a previous run() and stored in the state of the 9 | session. 10 | 11 | ##### Args: 12 | 13 | 14 | * `dtype`: The type of the output tensor. 15 | * `name`: Optional name prefix for the return tensor. 16 | 17 | ##### Returns: 18 | 19 | A pair of tensors. The first is a placeholder for feeding a 20 | tensor handle and the second is the tensor in the session state 21 | keyed by the tensor handle. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.ifft.md: -------------------------------------------------------------------------------- 1 | ### `tf.ifft(input, name=None)` {#ifft} 2 | 3 | Compute the inverse 1-dimensional discrete Fourier Transform. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor` of type `complex64`. A complex64 vector. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `complex64`. 14 | The inverse 1D Fourier Transform of `input`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.image.random_flip_left_right.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.random_flip_left_right(image, seed=None)` {#random_flip_left_right} 2 | 3 | Randomly flip an image horizontally (left to right). 4 | 5 | With a 1 in 2 chance, outputs the contents of `image` flipped along the 6 | second dimension, which is `width`. Otherwise output the image as-is. 7 | 8 | ##### Args: 9 | 10 | 11 | * `image`: A 3-D tensor of shape `[height, width, channels].` 12 | * `seed`: A Python integer. Used to create a random seed. See 13 | [`set_random_seed`](../../api_docs/python/constant_op.md#set_random_seed) 14 | for behavior. 15 | 16 | ##### Returns: 17 | 18 | A 3-D tensor of the same type and shape as `image`. 19 | 20 | ##### Raises: 21 | 22 | 23 | * `ValueError`: if the shape of `image` not supported. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.less.md: -------------------------------------------------------------------------------- 1 | ### `tf.less(x, y, name=None)` {#less} 2 | 3 | Returns the truth value of (x < y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.lgamma.md: -------------------------------------------------------------------------------- 1 | ### `tf.lgamma(x, name=None)` {#lgamma} 2 | 3 | Computes the log of the absolute value of `Gamma(x)` element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.logical_or.md: -------------------------------------------------------------------------------- 1 | ### `tf.logical_or(x, y, name=None)` {#logical_or} 2 | 3 | Returns the truth value of x OR y element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` of type `bool`. 9 | * `y`: A `Tensor` of type `bool`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.matrix_solve.md: -------------------------------------------------------------------------------- 1 | ### `tf.matrix_solve(matrix, rhs, adjoint=None, name=None)` {#matrix_solve} 2 | 3 | Solves a system of linear equations. Checks for invertibility. 4 | 5 | ##### Args: 6 | 7 | 8 | * `matrix`: A `Tensor`. Must be one of the following types: `float64`, `float32`. 9 | Shape is `[M, M]`. 10 | * `rhs`: A `Tensor`. Must have the same type as `matrix`. Shape is `[M, K]`. 11 | * `adjoint`: An optional `bool`. Defaults to `False`. 12 | Boolean indicating whether to solve with `matrix` or its adjoint. 13 | * `name`: A name for the operation (optional). 14 | 15 | ##### Returns: 16 | 17 | A `Tensor`. Has the same type as `matrix`. 18 | Shape is `[M, K]`. If `adjoint` is `False` then `output` that solves 19 | `matrix` * `output` = `rhs`. If `adjoint` is `True` then `output` that solves 20 | `adjoint(matrix)` * `output` = `rhs`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.ones_like.md: -------------------------------------------------------------------------------- 1 | ### `tf.ones_like(tensor, dtype=None, name=None)` {#ones_like} 2 | 3 | Creates a tensor with all elements set to 1. 4 | 5 | Given a single tensor (`tensor`), this operation returns a tensor of the same 6 | type and shape as `tensor` with all elements set to 1. Optionally, you can 7 | specify a new type (`dtype`) for the returned tensor. 8 | 9 | For example: 10 | 11 | ```python 12 | # 'tensor' is [[1, 2, 3], [4, 5, 6]] 13 | tf.ones_like(tensor) ==> [[1, 1, 1], [1, 1, 1]] 14 | ``` 15 | 16 | ##### Args: 17 | 18 | 19 | * `tensor`: A `Tensor`. 20 | * `dtype`: A type for the returned `Tensor`. Must be `float32`, `float64`, 21 | `int8`, `int16`, `int32`, `int64`, `uint8`, `complex64`, or `complex128`. 22 | 23 | * `name`: A name for the operation (optional). 24 | 25 | ##### Returns: 26 | 27 | A `Tensor` with all elements set to 1. 28 | 29 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.polygamma.md: -------------------------------------------------------------------------------- 1 | ### `tf.polygamma(a, x, name=None)` {#polygamma} 2 | 3 | Compute the polygamma function \\(\psi^{(n)}(x)\\). 4 | 5 | The polygamma function is defined as: 6 | 7 | ``` 8 | \psi^{(n)}(x) = \frac{d^n}{dx^n} \psi(x) 9 | ``` 10 | where \\(\psi(x)\\) is the digamma function. 11 | 12 | ##### Args: 13 | 14 | 15 | * `a`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 16 | * `x`: A `Tensor`. Must have the same type as `a`. 17 | * `name`: A name for the operation (optional). 18 | 19 | ##### Returns: 20 | 21 | A `Tensor`. Has the same type as `a`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.square.md: -------------------------------------------------------------------------------- 1 | ### `tf.square(x, name=None)` {#square} 2 | 3 | Computes square of x element-wise. 4 | 5 | I.e., \\(y = x * x = x^2\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.string_to_hash_bucket.md: -------------------------------------------------------------------------------- 1 | ### `tf.string_to_hash_bucket(string_tensor, num_buckets, name=None)` {#string_to_hash_bucket} 2 | 3 | Converts each string in the input Tensor to its hash mod by a number of buckets. 4 | 5 | The hash function is deterministic on the content of the string within the 6 | process. 7 | 8 | Note that the hash function may change from time to time. 9 | 10 | ##### Args: 11 | 12 | 13 | * `string_tensor`: A `Tensor` of type `string`. 14 | * `num_buckets`: An `int` that is `>= 1`. The number of buckets. 15 | * `name`: A name for the operation (optional). 16 | 17 | ##### Returns: 18 | 19 | A `Tensor` of type `int64`. 20 | A Tensor of the same shape as the input `string_tensor`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.train.MomentumOptimizer.md: -------------------------------------------------------------------------------- 1 | Optimizer that implements the Momentum algorithm. 2 | 3 | - - - 4 | 5 | #### `tf.train.MomentumOptimizer.__init__(learning_rate, momentum, use_locking=False, name='Momentum')` {#MomentumOptimizer.__init__} 6 | 7 | Construct a new Momentum optimizer. 8 | 9 | ##### Args: 10 | 11 | 12 | * `learning_rate`: A `Tensor` or a floating point value. The learning rate. 13 | * `momentum`: A `Tensor` or a floating point value. The momentum. 14 | * `use_locking`: If `True` use locks for update operations. 15 | * `name`: Optional name prefix for the operations created when applying 16 | gradients. Defaults to "Momentum". 17 | 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard7/tf.train.get_checkpoint_state.md: -------------------------------------------------------------------------------- 1 | ### `tf.train.get_checkpoint_state(checkpoint_dir, latest_filename=None)` {#get_checkpoint_state} 2 | 3 | Returns CheckpointState proto from the "checkpoint" file. 4 | 5 | If the "checkpoint" file contains a valid CheckpointState 6 | proto, returns it. 7 | 8 | ##### Args: 9 | 10 | 11 | * `checkpoint_dir`: The directory of checkpoints. 12 | * `latest_filename`: Optional name of the checkpoint file. Default to 13 | 'checkpoint'. 14 | 15 | ##### Returns: 16 | 17 | A CheckpointState if the state was available, None 18 | otherwise. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.VarLenFeature.md: -------------------------------------------------------------------------------- 1 | Configuration for parsing a variable-length input feature. 2 | 3 | Fields: 4 | dtype: Data type of input. 5 | - - - 6 | 7 | #### `tf.VarLenFeature.dtype` {#VarLenFeature.dtype} 8 | 9 | Alias for field number 0 10 | 11 | 12 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.acos.md: -------------------------------------------------------------------------------- 1 | ### `tf.acos(x, name=None)` {#acos} 2 | 3 | Computes acos of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.argmax.md: -------------------------------------------------------------------------------- 1 | ### `tf.argmax(input, dimension, name=None)` {#argmax} 2 | 3 | Returns the index with the largest value across dimensions of a tensor. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int64`, `int32`, `uint8`, `uint16`, `int16`, `int8`, `complex64`, `complex128`, `qint8`, `quint8`, `qint32`, `half`. 9 | * `dimension`: A `Tensor` of type `int32`. 10 | int32, 0 <= dimension < rank(input). Describes which dimension 11 | of the input Tensor to reduce across. For vectors, use dimension = 0. 12 | * `name`: A name for the operation (optional). 13 | 14 | ##### Returns: 15 | 16 | A `Tensor` of type `int64`. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.assert_proper_iterable.md: -------------------------------------------------------------------------------- 1 | ### `tf.assert_proper_iterable(values)` {#assert_proper_iterable} 2 | 3 | Static assert that values is a "proper" iterable. 4 | 5 | `Ops` that expect iterables of `Tensor` can call this to validate input. 6 | Useful since `Tensor`, `ndarray`, byte/text type are all iterables themselves. 7 | 8 | ##### Args: 9 | 10 | 11 | * `values`: Object to be checked. 12 | 13 | ##### Raises: 14 | 15 | 16 | * `TypeError`: If `values` is not iterable or is one of 17 | `Tensor`, `SparseTensor`, `np.array`, `tf.compat.bytes_or_text_types`. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.atan.md: -------------------------------------------------------------------------------- 1 | ### `tf.atan(x, name=None)` {#atan} 2 | 3 | Computes atan of x element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor`. Has the same type as `x`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.layers.summarize_tensor.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.summarize_tensor(tensor, tag=None)` {#summarize_tensor} 2 | 3 | Summarize a tensor using a suitable summary type. 4 | 5 | This function adds a summary op for `tensor`. The type of summary depends on 6 | the shape of `tensor`. For scalars, a `scalar_summary` is created, for all 7 | other tensors, `histogram_summary` is used. 8 | 9 | ##### Args: 10 | 11 | 12 | * `tensor`: The tensor to summarize 13 | * `tag`: The tag to use, if None then use tensor's op's name. 14 | 15 | ##### Returns: 16 | 17 | The summary op created or None for string tensors. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.learn.NanLossDuringTrainingError.md: -------------------------------------------------------------------------------- 1 | 2 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.contrib.learn.extract_pandas_labels.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.extract_pandas_labels(labels)` {#extract_pandas_labels} 2 | 3 | Extract data from pandas.DataFrame for labels. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.cross.md: -------------------------------------------------------------------------------- 1 | ### `tf.cross(a, b, name=None)` {#cross} 2 | 3 | Compute the pairwise cross product. 4 | 5 | `a` and `b` must be the same shape; they can either be simple 3-element vectors, 6 | or any shape where the innermost dimension is 3. In the latter case, each pair 7 | of corresponding 3-element vectors is cross-multiplied independently. 8 | 9 | ##### Args: 10 | 11 | 12 | * `a`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 13 | A tensor containing 3-element vectors. 14 | * `b`: A `Tensor`. Must have the same type as `a`. 15 | Another tensor, of same type and shape as `a`. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor`. Has the same type as `a`. 21 | Pairwise cross product of the vectors in `a` and `b`. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.equal.md: -------------------------------------------------------------------------------- 1 | ### `tf.equal(x, y, name=None)` {#equal} 2 | 3 | Returns the truth value of (x == y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `uint8`, `int8`, `int16`, `int32`, `int64`, `complex64`, `quint8`, `qint8`, `qint32`, `string`, `bool`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.image.transpose_image.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.transpose_image(image)` {#transpose_image} 2 | 3 | Transpose an image by swapping the first and second dimension. 4 | 5 | See also `transpose()`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `image`: 3-D tensor of shape `[height, width, channels]` 11 | 12 | ##### Returns: 13 | 14 | A 3-D tensor of shape `[width, height, channels]` 15 | 16 | ##### Raises: 17 | 18 | 19 | * `ValueError`: if the shape of `image` not supported. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.is_inf.md: -------------------------------------------------------------------------------- 1 | ### `tf.is_inf(x, name=None)` {#is_inf} 2 | 3 | Returns which elements of x are Inf. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` of type `bool`. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.less_equal.md: -------------------------------------------------------------------------------- 1 | ### `tf.less_equal(x, y, name=None)` {#less_equal} 2 | 3 | Returns the truth value of (x <= y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `float32`, `float64`, `int32`, `int64`, `uint8`, `int16`, `int8`, `uint16`, `half`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` of type `bool`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.relu6.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.relu6(features, name=None)` {#relu6} 2 | 3 | Computes Rectified Linear 6: `min(max(features, 0), 6)`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `features`: A `Tensor` with type `float`, `double`, `int32`, `int64`, `uint8`, 9 | `int16`, or `int8`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor` with the same type as `features`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.nn.rnn_cell.LSTMStateTuple.md: -------------------------------------------------------------------------------- 1 | Tuple used by LSTM Cells for `state_size`, `zero_state`, and output state. 2 | 3 | Stores two elements: `(c, h)`, in that order. 4 | 5 | Only used when `state_is_tuple=True`. 6 | - - - 7 | 8 | #### `tf.nn.rnn_cell.LSTMStateTuple.c` {#LSTMStateTuple.c} 9 | 10 | Alias for field number 0 11 | 12 | 13 | - - - 14 | 15 | #### `tf.nn.rnn_cell.LSTMStateTuple.h` {#LSTMStateTuple.h} 16 | 17 | Alias for field number 1 18 | 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.no_regularizer.md: -------------------------------------------------------------------------------- 1 | ### `tf.no_regularizer(_)` {#no_regularizer} 2 | 3 | Use this function to prevent regularization of variables. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.ones_initializer.md: -------------------------------------------------------------------------------- 1 | ### `tf.ones_initializer(shape, dtype=tf.float32)` {#ones_initializer} 2 | 3 | An adaptor for ones() to match the Initializer spec. 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.reset_default_graph.md: -------------------------------------------------------------------------------- 1 | ### `tf.reset_default_graph()` {#reset_default_graph} 2 | 3 | Clears the default graph stack and resets the global default graph. 4 | 5 | NOTE: The default graph is a property of the current thread. This 6 | function applies only to the current thread. Calling this function while 7 | a `tf.Session` or `tf.InteractiveSession` is active will result in undefined 8 | behavior. Using any previously created `tf.Operation` or `tf.Tensor` objects 9 | after calling this function will result in undefined behavior. 10 | 11 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.round.md: -------------------------------------------------------------------------------- 1 | ### `tf.round(x, name=None)` {#round} 2 | 3 | Rounds the values of a tensor to the nearest integer, element-wise. 4 | 5 | For example: 6 | 7 | ```python 8 | # 'a' is [0.9, 2.5, 2.3, -4.4] 9 | tf.round(a) ==> [ 1.0, 3.0, 2.0, -4.0 ] 10 | ``` 11 | 12 | ##### Args: 13 | 14 | 15 | * `x`: A `Tensor` of type `float` or `double`. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor` of same shape and type as `x`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.rsqrt.md: -------------------------------------------------------------------------------- 1 | ### `tf.rsqrt(x, name=None)` {#rsqrt} 2 | 3 | Computes reciprocal of square root of x element-wise. 4 | 5 | I.e., \\(y = 1 / \sqrt{x}\\). 6 | 7 | ##### Args: 8 | 9 | 10 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `complex64`, `complex128`. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor`. Has the same type as `x`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard8/tf.zeta.md: -------------------------------------------------------------------------------- 1 | ### `tf.zeta(x, q, name=None)` {#zeta} 2 | 3 | Compute the Hurwitz zeta function \\(\zeta(x, q)\\). 4 | 5 | The Hurwitz zeta function is defined as: 6 | 7 | ``` 8 | \zeta(x, q) = \sum_{n=0}^{\infty} (q + n)^{-x} 9 | ``` 10 | 11 | ##### Args: 12 | 13 | 14 | * `x`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 15 | * `q`: A `Tensor`. Must have the same type as `x`. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | A `Tensor`. Has the same type as `x`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.batch_fft2d.md: -------------------------------------------------------------------------------- 1 | ### `tf.batch_fft2d(input, name=None)` {#batch_fft2d} 2 | 3 | Compute the 2-dimensional discrete Fourier Transform over the inner-most 4 | 5 | 2 dimensions of `input`. 6 | 7 | ##### Args: 8 | 9 | 10 | * `input`: A `Tensor` of type `complex64`. A complex64 tensor. 11 | * `name`: A name for the operation (optional). 12 | 13 | ##### Returns: 14 | 15 | A `Tensor` of type `complex64`. 16 | A complex64 tensor of the same shape as `input`. The inner-most 2 17 | dimensions of `input` are replaced with their 2D Fourier Transform. 18 | 19 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.framework.get_or_create_global_step.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.framework.get_or_create_global_step(graph=None)` {#get_or_create_global_step} 2 | 3 | Returns and create (if necessary) the global step variable. 4 | 5 | ##### Args: 6 | 7 | 8 | * `graph`: The graph in which to create the global step. If missing, use default 9 | graph. 10 | 11 | ##### Returns: 12 | 13 | the tensor representing the global step variable. 14 | 15 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.layers.l1_regularizer.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.layers.l1_regularizer(scale)` {#l1_regularizer} 2 | 3 | Returns a function that can be used to apply L1 regularization to weights. 4 | 5 | L1 regularization encourages sparsity. 6 | 7 | ##### Args: 8 | 9 | 10 | * `scale`: A scalar multiplier `Tensor`. 0.0 disables the regularizer. 11 | 12 | ##### Returns: 13 | 14 | A function with signature `l1(weights, name=None)` that apply L1 15 | regularization. 16 | 17 | ##### Raises: 18 | 19 | 20 | * `ValueError`: If scale is outside of the range [0.0, 1.0] or if scale is not a 21 | float. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.learn.infer.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.learn.infer(restore_checkpoint_path, output_dict, feed_dict=None)` {#infer} 2 | 3 | 4 | 5 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.contrib.losses.get_total_loss.md: -------------------------------------------------------------------------------- 1 | ### `tf.contrib.losses.get_total_loss(add_regularization_losses=True, name='total_loss')` {#get_total_loss} 2 | 3 | Returns a tensor whose value represents the total loss. 4 | 5 | Notice that the function adds the given losses to the regularization losses. 6 | 7 | ##### Args: 8 | 9 | 10 | * `add_regularization_losses`: A boolean indicating whether or not to use the 11 | regularization losses in the sum. 12 | * `name`: The name of the returned tensor. 13 | 14 | ##### Returns: 15 | 16 | A `Tensor` whose value represents the total loss. 17 | 18 | ##### Raises: 19 | 20 | 21 | * `ValueError`: if `losses` is not iterable. 22 | 23 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.errors.InvalidArgumentError.md: -------------------------------------------------------------------------------- 1 | Raised when an operation receives an invalid argument. 2 | 3 | This may occur, for example, if an operation is receives an input 4 | tensor that has an invalid value or shape. For example, the 5 | [`tf.matmul()`](../../api_docs/python/math_ops.md#matmul) op will raise this 6 | error if it receives an input that is not a matrix, and the 7 | [`tf.reshape()`](../../api_docs/python/array_ops.md#reshape) op will raise 8 | this error if the new shape does not match the number of elements in the input 9 | tensor. 10 | 11 | - - - 12 | 13 | #### `tf.errors.InvalidArgumentError.__init__(node_def, op, message)` {#InvalidArgumentError.__init__} 14 | 15 | Creates an `InvalidArgumentError`. 16 | 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.errors.UnknownError.md: -------------------------------------------------------------------------------- 1 | Unknown error. 2 | 3 | An example of where this error may be returned is if a Status value 4 | received from another address space belongs to an error-space that 5 | is not known to this address space. Also errors raised by APIs that 6 | do not return enough error information may be converted to this 7 | error. 8 | 9 | - - - 10 | 11 | #### `tf.errors.UnknownError.__init__(node_def, op, message, error_code=2)` {#UnknownError.__init__} 12 | 13 | Creates an `UnknownError`. 14 | 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.fill.md: -------------------------------------------------------------------------------- 1 | ### `tf.fill(dims, value, name=None)` {#fill} 2 | 3 | Creates a tensor filled with a scalar value. 4 | 5 | This operation creates a tensor of shape `dims` and fills it with `value`. 6 | 7 | For example: 8 | 9 | ```prettyprint 10 | # Output tensor has shape [2, 3]. 11 | fill([2, 3], 9) ==> [[9, 9, 9] 12 | [9, 9, 9]] 13 | ``` 14 | 15 | ##### Args: 16 | 17 | 18 | * `dims`: A `Tensor` of type `int32`. 19 | 1-D. Represents the shape of the output tensor. 20 | * `value`: A `Tensor`. 0-D (scalar). Value to fill the returned tensor. 21 | * `name`: A name for the operation (optional). 22 | 23 | ##### Returns: 24 | 25 | A `Tensor`. Has the same type as `value`. 26 | 27 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.get_default_graph.md: -------------------------------------------------------------------------------- 1 | ### `tf.get_default_graph()` {#get_default_graph} 2 | 3 | Returns the default graph for the current thread. 4 | 5 | The returned graph will be the innermost graph on which a 6 | `Graph.as_default()` context has been entered, or a global default 7 | graph if none has been explicitly created. 8 | 9 | NOTE: The default graph is a property of the current thread. If you 10 | create a new thread, and wish to use the default graph in that 11 | thread, you must explicitly add a `with g.as_default():` in that 12 | thread's function. 13 | 14 | ##### Returns: 15 | 16 | The default `Graph` being used in the current thread. 17 | 18 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.image.random_flip_up_down.md: -------------------------------------------------------------------------------- 1 | ### `tf.image.random_flip_up_down(image, seed=None)` {#random_flip_up_down} 2 | 3 | Randomly flips an image vertically (upside down). 4 | 5 | With a 1 in 2 chance, outputs the contents of `image` flipped along the first 6 | dimension, which is `height`. Otherwise output the image as-is. 7 | 8 | ##### Args: 9 | 10 | 11 | * `image`: A 3-D tensor of shape `[height, width, channels].` 12 | * `seed`: A Python integer. Used to create a random seed. See 13 | [`set_random_seed`](../../api_docs/python/constant_op.md#set_random_seed) 14 | for behavior. 15 | 16 | ##### Returns: 17 | 18 | A 3-D tensor of the same type and shape as `image`. 19 | 20 | ##### Raises: 21 | 22 | 23 | * `ValueError`: if the shape of `image` not supported. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.matrix_determinant.md: -------------------------------------------------------------------------------- 1 | ### `tf.matrix_determinant(input, name=None)` {#matrix_determinant} 2 | 3 | Calculates the determinant of a square matrix. 4 | 5 | ##### Args: 6 | 7 | 8 | * `input`: A `Tensor`. Must be one of the following types: `float32`, `float64`. 9 | A tensor of shape `[M, M]`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `input`. 15 | A scalar, equal to the determinant of the input. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.nn.zero_fraction.md: -------------------------------------------------------------------------------- 1 | ### `tf.nn.zero_fraction(value, name=None)` {#zero_fraction} 2 | 3 | Returns the fraction of zeros in `value`. 4 | 5 | If `value` is empty, the result is `nan`. 6 | 7 | This is useful in summaries to measure and report sparsity. For example, 8 | 9 | z = tf.Relu(...) 10 | summ = tf.scalar_summary('sparsity', tf.nn.zero_fraction(z)) 11 | 12 | ##### Args: 13 | 14 | 15 | * `value`: A tensor of numeric type. 16 | * `name`: A name for the operation (optional). 17 | 18 | ##### Returns: 19 | 20 | The fraction of zeros in `value`, with type `float32`. 21 | 22 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.ones.md: -------------------------------------------------------------------------------- 1 | ### `tf.ones(shape, dtype=tf.float32, name=None)` {#ones} 2 | 3 | Creates a tensor with all elements set to 1. 4 | 5 | This operation returns a tensor of type `dtype` with shape `shape` and all 6 | elements set to 1. 7 | 8 | For example: 9 | 10 | ```python 11 | tf.ones([2, 3], int32) ==> [[1, 1, 1], [1, 1, 1]] 12 | ``` 13 | 14 | ##### Args: 15 | 16 | 17 | * `shape`: Either a list of integers, or a 1-D `Tensor` of type `int32`. 18 | * `dtype`: The type of an element in the resulting `Tensor`. 19 | * `name`: A name for the operation (optional). 20 | 21 | ##### Returns: 22 | 23 | A `Tensor` with all elements set to 1. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.saturate_cast.md: -------------------------------------------------------------------------------- 1 | ### `tf.saturate_cast(value, dtype, name=None)` {#saturate_cast} 2 | 3 | Performs a safe saturating cast of `value` to `dtype`. 4 | 5 | This function casts the input to `dtype` without applying any scaling. If 6 | there is a danger that values would over or underflow in the cast, this op 7 | applies the appropriate clamping before the cast. 8 | 9 | ##### Args: 10 | 11 | 12 | * `value`: A `Tensor`. 13 | * `dtype`: The desired output `DType`. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | `value` safely cast to `dtype`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.scalar_mul.md: -------------------------------------------------------------------------------- 1 | ### `tf.scalar_mul(scalar, x)` {#scalar_mul} 2 | 3 | Multiplies a scalar times a `Tensor` or `IndexedSlices` object. 4 | 5 | Intended for use in gradient code which might deal with `IndexedSlices` 6 | objects, which are easy to multiply by a scalar but more expensive to 7 | multiply with arbitrary tensors. 8 | 9 | ##### Args: 10 | 11 | 12 | * `scalar`: A 0-D scalar `Tensor`. Must have known shape. 13 | * `x`: A `Tensor` or `IndexedSlices` to be scaled. 14 | 15 | ##### Returns: 16 | 17 | `scalar * x` of the same type (`Tensor` or `IndexedSlices`) as `x`. 18 | 19 | ##### Raises: 20 | 21 | 22 | * `ValueError`: if scalar is not a 0-D `scalar`. 23 | 24 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.size.md: -------------------------------------------------------------------------------- 1 | ### `tf.size(input, name=None)` {#size} 2 | 3 | Returns the size of a tensor. 4 | 5 | This operation returns an integer representing the number of elements in 6 | `input`. 7 | 8 | For example: 9 | 10 | ```prettyprint 11 | # 't' is [[[1, 1,, 1], [2, 2, 2]], [[3, 3, 3], [4, 4, 4]]]] 12 | size(t) ==> 12 13 | ``` 14 | 15 | ##### Args: 16 | 17 | 18 | * `input`: A `Tensor`. 19 | * `name`: A name for the operation (optional). 20 | 21 | ##### Returns: 22 | 23 | A `Tensor` of type `int32`. 24 | 25 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.squared_difference.md: -------------------------------------------------------------------------------- 1 | ### `tf.squared_difference(x, y, name=None)` {#squared_difference} 2 | 3 | Returns (x - y)(x - y) element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor`. Must be one of the following types: `half`, `float32`, `float64`, `int32`, `int64`, `complex64`, `complex128`. 9 | * `y`: A `Tensor`. Must have the same type as `x`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A `Tensor`. Has the same type as `x`. 15 | 16 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.string_to_number.md: -------------------------------------------------------------------------------- 1 | ### `tf.string_to_number(string_tensor, out_type=None, name=None)` {#string_to_number} 2 | 3 | Converts each string in the input Tensor to the specified numeric type. 4 | 5 | (Note that int32 overflow results in an error while float overflow 6 | results in a rounded value.) 7 | 8 | ##### Args: 9 | 10 | 11 | * `string_tensor`: A `Tensor` of type `string`. 12 | * `out_type`: An optional `tf.DType` from: `tf.float32, tf.int32`. Defaults to `tf.float32`. 13 | The numeric type to interpret each string in string_tensor as. 14 | * `name`: A name for the operation (optional). 15 | 16 | ##### Returns: 17 | 18 | A `Tensor` of type `out_type`. 19 | A Tensor of the same shape as the input `string_tensor`. 20 | 21 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.tanh.md: -------------------------------------------------------------------------------- 1 | ### `tf.tanh(x, name=None)` {#tanh} 2 | 3 | Computes hyperbolic tangent of `x` element-wise. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A Tensor with type `float`, `double`, `int32`, `complex64`, `int64`, 9 | or `qint32`. 10 | * `name`: A name for the operation (optional). 11 | 12 | ##### Returns: 13 | 14 | A Tensor with the same type as `x` if `x.dtype != qint32` otherwise 15 | the return type is `quint8`. 16 | 17 | -------------------------------------------------------------------------------- /g3doc/api_docs/python/functions_and_classes/shard9/tf.to_bfloat16.md: -------------------------------------------------------------------------------- 1 | ### `tf.to_bfloat16(x, name='ToBFloat16')` {#to_bfloat16} 2 | 3 | Casts a tensor to type `bfloat16`. 4 | 5 | ##### Args: 6 | 7 | 8 | * `x`: A `Tensor` or `SparseTensor`. 9 | * `name`: A name for the operation (optional). 10 | 11 | ##### Returns: 12 | 13 | A `Tensor` or `SparseTensor` with same shape as `x` with type `bfloat16`. 14 | 15 | ##### Raises: 16 | 17 | 18 | * `TypeError`: If `x` cannot be cast to the `bfloat16`. 19 | 20 | -------------------------------------------------------------------------------- /g3doc/extras/README.txt: -------------------------------------------------------------------------------- 1 | This directory holds extra files we'd like to be able 2 | to link to and serve from within tensorflow.org. 3 | They are excluded from versioning. -------------------------------------------------------------------------------- /g3doc/get_started/leftnav_files: -------------------------------------------------------------------------------- 1 | index.md 2 | os_setup.md 3 | basic_usage.md -------------------------------------------------------------------------------- /g3doc/how_tos/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/how_tos/__init__.py -------------------------------------------------------------------------------- /g3doc/how_tos/adding_an_op/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/how_tos/adding_an_op/__init__.py -------------------------------------------------------------------------------- /g3doc/how_tos/leftnav_files: -------------------------------------------------------------------------------- 1 | variables/index.md 2 | ../tutorials/mnist/tf/index.md 3 | summaries_and_tensorboard/index.md 4 | graph_viz/index.md 5 | reading_data/index.md 6 | threading_and_queues/index.md 7 | distributed/index.md 8 | adding_an_op/index.md 9 | new_data_formats/index.md 10 | using_gpu/index.md 11 | variable_scope/index.md -------------------------------------------------------------------------------- /g3doc/images/AlexClassification.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/AlexClassification.png -------------------------------------------------------------------------------- /g3doc/images/AnimatedFileQueues.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/AnimatedFileQueues.gif -------------------------------------------------------------------------------- /g3doc/images/DynamicPartition.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/DynamicPartition.png -------------------------------------------------------------------------------- /g3doc/images/DynamicStitch.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/DynamicStitch.png -------------------------------------------------------------------------------- /g3doc/images/Gather.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/Gather.png -------------------------------------------------------------------------------- /g3doc/images/IncremeterFifoQueue.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/IncremeterFifoQueue.gif -------------------------------------------------------------------------------- /g3doc/images/MNIST-Matrix.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/MNIST-Matrix.png -------------------------------------------------------------------------------- /g3doc/images/MNIST.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/MNIST.png -------------------------------------------------------------------------------- /g3doc/images/Parallelism.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/Parallelism.png -------------------------------------------------------------------------------- /g3doc/images/audio-image-text.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/audio-image-text.png -------------------------------------------------------------------------------- /g3doc/images/baseball_network.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/baseball_network.png -------------------------------------------------------------------------------- /g3doc/images/blue_pill.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/blue_pill.png -------------------------------------------------------------------------------- /g3doc/images/cifar_activations.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_activations.png -------------------------------------------------------------------------------- /g3doc/images/cifar_graph.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_graph.png -------------------------------------------------------------------------------- /g3doc/images/cifar_image_summary.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_image_summary.png -------------------------------------------------------------------------------- /g3doc/images/cifar_loss.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_loss.png -------------------------------------------------------------------------------- /g3doc/images/cifar_lr_decay.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_lr_decay.png -------------------------------------------------------------------------------- /g3doc/images/cifar_samples.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_samples.png -------------------------------------------------------------------------------- /g3doc/images/cifar_sparsity.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_sparsity.png -------------------------------------------------------------------------------- /g3doc/images/cifar_var_histograms.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cifar_var_histograms.png -------------------------------------------------------------------------------- /g3doc/images/colorby_compute_time.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/colorby_compute_time.png -------------------------------------------------------------------------------- /g3doc/images/colorby_device.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/colorby_device.png -------------------------------------------------------------------------------- /g3doc/images/colorby_structure.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/colorby_structure.png -------------------------------------------------------------------------------- /g3doc/images/constant.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/constant.png -------------------------------------------------------------------------------- /g3doc/images/control_edge.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/control_edge.png -------------------------------------------------------------------------------- /g3doc/images/conv_1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/conv_1.png -------------------------------------------------------------------------------- /g3doc/images/cropped_panda.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/cropped_panda.jpg -------------------------------------------------------------------------------- /g3doc/images/dataflow_edge.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/dataflow_edge.png -------------------------------------------------------------------------------- /g3doc/images/getting_started.dot: -------------------------------------------------------------------------------- 1 | digraph Dependencies { 2 | node [shape = oval]; 3 | "predictions: MatMul()" -> "data: Concat()" 4 | "data: Concat()" -> data_left 5 | "data: Concat()" -> data_right 6 | "predictions: MatMul()" -> "weight_matrix: Reshape()" 7 | "weight_matrix: Reshape()" -> "new_weights: Add()" 8 | "new_weights: Add()" -> weights 9 | "new_weights: Add()" -> deltas 10 | "update: Assign()" -> weights 11 | "update: Assign()" -> "new_weights: Add()" 12 | "InitializeAllVariables()" -> weights 13 | "InitializeAllVariables()" -> init_value 14 | } -------------------------------------------------------------------------------- /g3doc/images/getting_started.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/getting_started.png -------------------------------------------------------------------------------- /g3doc/images/grace_hopper.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/grace_hopper.jpg -------------------------------------------------------------------------------- /g3doc/images/graph_vis_animation.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/graph_vis_animation.gif -------------------------------------------------------------------------------- /g3doc/images/horizontal_stack.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/horizontal_stack.png -------------------------------------------------------------------------------- /g3doc/images/infocard.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/infocard.png -------------------------------------------------------------------------------- /g3doc/images/infocard_op.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/infocard_op.png -------------------------------------------------------------------------------- /g3doc/images/iris_three_species.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/iris_three_species.jpg -------------------------------------------------------------------------------- /g3doc/images/linear-relationships.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/linear-relationships.png -------------------------------------------------------------------------------- /g3doc/images/mandelbrot_output.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mandelbrot_output.jpg -------------------------------------------------------------------------------- /g3doc/images/mnist-train-xs.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist-train-xs.png -------------------------------------------------------------------------------- /g3doc/images/mnist-train-ys.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist-train-ys.png -------------------------------------------------------------------------------- /g3doc/images/mnist1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist1.png -------------------------------------------------------------------------------- /g3doc/images/mnist10.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist10.png -------------------------------------------------------------------------------- /g3doc/images/mnist2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist2.png -------------------------------------------------------------------------------- /g3doc/images/mnist3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist3.png -------------------------------------------------------------------------------- /g3doc/images/mnist4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist4.png -------------------------------------------------------------------------------- /g3doc/images/mnist5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist5.png -------------------------------------------------------------------------------- /g3doc/images/mnist6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist6.png -------------------------------------------------------------------------------- /g3doc/images/mnist7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist7.png -------------------------------------------------------------------------------- /g3doc/images/mnist8.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist8.png -------------------------------------------------------------------------------- /g3doc/images/mnist9.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist9.png -------------------------------------------------------------------------------- /g3doc/images/mnist_digits.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist_digits.png -------------------------------------------------------------------------------- /g3doc/images/mnist_subgraph.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist_subgraph.png -------------------------------------------------------------------------------- /g3doc/images/mnist_tensorboard.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/mnist_tensorboard.png -------------------------------------------------------------------------------- /g3doc/images/namespace_node.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/namespace_node.png -------------------------------------------------------------------------------- /g3doc/images/nce-nplm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/nce-nplm.png -------------------------------------------------------------------------------- /g3doc/images/op_node.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/op_node.png -------------------------------------------------------------------------------- /g3doc/images/pde_output_1.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/pde_output_1.jpg -------------------------------------------------------------------------------- /g3doc/images/pde_output_2.jpg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/pde_output_2.jpg -------------------------------------------------------------------------------- /g3doc/images/pool1_collapsed.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/pool1_collapsed.png -------------------------------------------------------------------------------- /g3doc/images/pool1_expanded.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/pool1_expanded.png -------------------------------------------------------------------------------- /g3doc/images/re.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/re.png -------------------------------------------------------------------------------- /g3doc/images/re1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/re1.png -------------------------------------------------------------------------------- /g3doc/images/red_pill.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/red_pill.png -------------------------------------------------------------------------------- /g3doc/images/reference_edge.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/reference_edge.png -------------------------------------------------------------------------------- /g3doc/images/results.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/results.png -------------------------------------------------------------------------------- /g3doc/images/run_metadata_graph.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/run_metadata_graph.png -------------------------------------------------------------------------------- /g3doc/images/run_metadata_infocard.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/run_metadata_infocard.png -------------------------------------------------------------------------------- /g3doc/images/rw4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/rw4.png -------------------------------------------------------------------------------- /g3doc/images/rw5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/rw5.png -------------------------------------------------------------------------------- /g3doc/images/rw6.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/rw6.png -------------------------------------------------------------------------------- /g3doc/images/rw7.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/rw7.png -------------------------------------------------------------------------------- /g3doc/images/save.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/save.png -------------------------------------------------------------------------------- /g3doc/images/scatterplot.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/scatterplot.png -------------------------------------------------------------------------------- /g3doc/images/series.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/series.png -------------------------------------------------------------------------------- /g3doc/images/series_expanded.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/series_expanded.png -------------------------------------------------------------------------------- /g3doc/images/softmax-nplm.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/softmax-nplm.png -------------------------------------------------------------------------------- /g3doc/images/softmax-regression-scalarequation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/softmax-regression-scalarequation.png -------------------------------------------------------------------------------- /g3doc/images/softmax-regression-scalargraph.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/softmax-regression-scalargraph.png -------------------------------------------------------------------------------- /g3doc/images/softmax-regression-vectorequation.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/softmax-regression-vectorequation.png -------------------------------------------------------------------------------- /g3doc/images/softmax-weights.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/softmax-weights.png -------------------------------------------------------------------------------- /g3doc/images/summary.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/summary.png -------------------------------------------------------------------------------- /g3doc/images/tensor_shapes.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/tensor_shapes.png -------------------------------------------------------------------------------- /g3doc/images/tensors_flowing.gif: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/tensors_flowing.gif -------------------------------------------------------------------------------- /g3doc/images/tf_logo.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/tf_logo.png -------------------------------------------------------------------------------- /g3doc/images/tf_logo_transp.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/tf_logo_transp.png -------------------------------------------------------------------------------- /g3doc/images/theta.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/theta.png -------------------------------------------------------------------------------- /g3doc/images/tsne.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/tsne.png -------------------------------------------------------------------------------- /g3doc/images/vertical_stack.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vertical_stack.png -------------------------------------------------------------------------------- /g3doc/images/vr1.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vr1.png -------------------------------------------------------------------------------- /g3doc/images/vr2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vr2.png -------------------------------------------------------------------------------- /g3doc/images/vr3.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vr3.png -------------------------------------------------------------------------------- /g3doc/images/vr4.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vr4.png -------------------------------------------------------------------------------- /g3doc/images/vr5.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/vr5.png -------------------------------------------------------------------------------- /g3doc/images/word2vec2.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/images/word2vec2.png -------------------------------------------------------------------------------- /g3doc/index.md: -------------------------------------------------------------------------------- 1 | # TensorFlow 2 | 3 | 4 | 5 | ## 소개 6 | 7 | 텐서플로우(TensorFlow™)는 데이터 플로우 그래프(Data flow graph)를 사용하여 수치 연산을 하는 오픈소스 소프트웨어 라이브러리입니다. 그래프의 노드(Node)는 수치 연산을 나타내고 엣지(edge)는 노드 사이를 이동하는 다차원 데이터 배열(텐서,tensor)를 나타냅니다. 유연한 아키텍처로 구성되어 있어 코드 수정없이 데스크탑, 서버 혹은 모바일 디바이스에서 CPU나 GPU를 사용하여 연산을 구동시킬 수 있습니다. 텐서플로우는 원래 머신러닝과 딥 뉴럴 네트워크 연구를 목적으로 구글의 인공지능 연구 조직인 구글 브레인 팀의 연구자와 엔지니어들에 의해 개발되었습니다. 하지만 이 시스템은 여러 다른 분야에도 충분히 적용될 수 있습니다. 다음 문서에서 어떻게 텐서플로우를 설치하고 사용하는지에 대해 설명하겠습니다. 8 | 9 | ## Table of Contents 10 | 11 | 12 | -------------------------------------------------------------------------------- /g3doc/resources/leftnav_files: -------------------------------------------------------------------------------- 1 | bib.md 2 | uses.md 3 | faq.md 4 | glossary.md 5 | dims_types.md 6 | versions.md 7 | roadmap.md -------------------------------------------------------------------------------- /g3doc/tutorials/BUILD: -------------------------------------------------------------------------------- 1 | # Description: 2 | # Top-level tutorials files 3 | 4 | package(default_visibility = ["//tensorflow:internal"]) 5 | 6 | licenses(["notice"]) # Apache 2.0 7 | 8 | exports_files(["LICENSE"]) 9 | 10 | filegroup( 11 | name = "all_files", 12 | srcs = glob( 13 | ["**/*"], 14 | exclude = [ 15 | "**/METADATA", 16 | "**/OWNERS", 17 | ], 18 | ), 19 | ) 20 | -------------------------------------------------------------------------------- /g3doc/tutorials/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/tensorflowkorea/tensorflow-kr/779d4e1fcaa2e3a6277d1b01804470eb228e8fb0/g3doc/tutorials/__init__.py -------------------------------------------------------------------------------- /g3doc/tutorials/deep_cnn/cifar_tensorboard.html: -------------------------------------------------------------------------------- 1 | 2 | 3 | 4 | TensorBoard Demo 5 | 6 | 7 | 17 | 18 | 19 | 20 | 21 | 22 | -------------------------------------------------------------------------------- /g3doc/tutorials/leftnav_files: -------------------------------------------------------------------------------- 1 | ### Basic Neural Networks 2 | mnist/beginners/index.md 3 | mnist/pros/index.md 4 | mnist/tf/index.md 5 | mnist/download/index.md 6 | ### Easy ML with tf.contrib.learn 7 | tflearn/index.md 8 | linear/overview.md 9 | wide/index.md 10 | wide_and_deep/index.md 11 | ### TensorFlow Serving 12 | tfserve/index.md 13 | ### Image Processing 14 | deep_cnn/index.md 15 | image_recognition/index.md 16 | ### Language and Sequence Processing 17 | word2vec/index.md 18 | recurrent/index.md 19 | seq2seq/index.md 20 | syntaxnet/index.md 21 | ### Non-ML Applications 22 | mandelbrot/index.md 23 | pdes/index.md 24 | -------------------------------------------------------------------------------- /g3doc/tutorials/syntaxnet/index.md: -------------------------------------------------------------------------------- 1 | # SyntaxNet 2 | (v0.9) 3 | 4 | ## 소개 5 | 6 | SyntaxNet는 텐서플로우를 위한 뉴럴 네트워크 방식의 자연어 처리 프레임워크입니다. 7 | 8 | ## SyntaxNet의 기본 튜토리얼 9 | 10 | 해당 [튜토리얼]( 11 | https://github.com/tensorflow/models/tree/master/syntaxnet#installation) 12 | 은 다음 항목들을 설명하고 있습니다: 13 | 14 | * SyntaxNet를 설치하는 방법 15 | * SyntaxNet 패키지에 포함된 Parsey McParseface 파서를 사용하는 방법 16 | * 사용자의 품사 태거를 훈련하는 방법 17 | * 사용자의 파서를 훈련하는 방법 -------------------------------------------------------------------------------- /g3doc/tutorials/tfserve/index.md: -------------------------------------------------------------------------------- 1 | # TensorFlow Serving 2 | 3 | ## Introduction 4 | 5 | 텐서플로우 서빙은 제품 환경을 위해 디자인 되었으며, 머신러닝 모델을 위해 유연하며 고성능의 서빙 시스템을 제공합니다. 6 | 텐서플로우 서빙을 통해 같은 서버 아키텍쳐와 API를 유지하는 동안, 새로운 알고리즘과 실험을 쉽게 배포할 수 있습니다. 7 | 8 | ## Basic Serving Tutorial 9 | 10 | 텐서플로우 서빙 페이지의 [basic tutorial](https://tensorflow.github.io/serving/serving_basic)을 통해 학습된 텐서플로우 모델을 어떻게 내보낼지 그리고 서버를 통해 모델을 어떻게 올릴 수 있는지에 대해 학습하실 수 있습니다. 11 | 12 | ## Advanced Serving Tutorial 13 | 14 | 텐서플로우 서빙 페이지의 [advanced tutorial](https://tensorflow.github.io/serving/serving_advanced)을 통해 동적으로 서버를 구축하는 방법을 배우고 최신 버전의 학습된 텐서플로우 모델을 경험하실 수 있습니다. 15 | 16 | ## Serving Inception Model Tutorial 17 | 18 | 텐서플로우 서빙 페이지의 [serving inception tutorial](https://tensorflow.github.io/serving/serving_inception)을 통해 초기 모델을 텐서플로우 서빙과 Kubernetes으로 올리는 방법을 학습하실 수 있습니다. 19 | --------------------------------------------------------------------------------