├── .gitignore ├── README.md ├── eval ├── README.md ├── evaluate_humaneval.py ├── evaluate_humaneval_plus.py ├── evaluate_mbpp.py ├── evaluate_mbpp_plus.py ├── run_humaneval.py ├── run_humaneval_plus.py ├── run_mbpp.py └── run_mbpp_plus.py ├── model_release ├── LICENSE ├── pyproject.toml └── src │ └── fairseq_t5 │ ├── __init__.py │ ├── configuration_fairseq_t5.py │ ├── convert_to_hf_checkpoint.py │ ├── fairseq_dictionary.py │ ├── modeling_fairseq_t5.py │ └── tokenization_ast_t5.py └── training ├── README.md ├── examples ├── .gitignore ├── __init__.py └── t5 │ └── config │ └── pl │ ├── base1m_rpe_10_2e-4_64gpus_bs8192.yaml │ └── base1m_rpe_mask25_ast10_lb5_ub100_op50_or50_2e-4_64gpus_bs8192.yaml ├── fairseq ├── __init__.py ├── ast_t5_dataset_fast.pyi ├── benchmark │ ├── __init__.py │ ├── dummy_dataset.py │ ├── dummy_lm.py │ ├── dummy_masked_lm.py │ ├── dummy_model.py │ └── dummy_mt.py ├── binarizer.py ├── checkpoint_utils.py ├── clib │ ├── cuda │ │ ├── ngram_repeat_block_cuda.cpp │ │ └── ngram_repeat_block_cuda_kernel.cu │ ├── libast │ │ ├── interface.cpp │ │ └── libast.hpp │ ├── libbase │ │ └── balanced_assignment.cpp │ ├── libbleu │ │ ├── libbleu.cpp │ │ └── module.cpp │ ├── libnat │ │ └── edit_dist.cpp │ └── libnat_cuda │ │ ├── binding.cpp │ │ ├── edit_dist.cu │ │ └── edit_dist.h ├── config │ ├── __init__.py │ ├── config.yaml │ └── model │ │ └── transformer_lm │ │ ├── transformer_lm_big.yaml │ │ ├── transformer_lm_gbw.yaml │ │ ├── transformer_lm_gpt.yaml │ │ ├── transformer_lm_gpt2_big.yaml │ │ ├── transformer_lm_gpt2_medium.yaml │ │ ├── transformer_lm_gpt2_small.yaml │ │ └── transformer_lm_wiki103.yaml ├── criterions │ ├── __init__.py │ ├── fairseq_criterion.py │ ├── label_smoothed_cross_entropy.py │ ├── masked_lm.py │ ├── model_criterion.py │ ├── sentence_prediction.py │ └── sentence_ranking.py ├── data │ ├── __init__.py │ ├── add_target_dataset.py │ ├── append_token_dataset.py │ ├── ast_t5_dataset.py │ ├── base_wrapper_dataset.py │ ├── bucket_pad_length_dataset.py │ ├── colorize_dataset.py │ ├── concat_capped_dataset.py │ ├── concat_dataset.py │ ├── concat_sentences_dataset.py │ ├── concat_weighted_dataset.py │ ├── constant_dataset.py │ ├── data_utils.py │ ├── data_utils_fast.pyx │ ├── dictionary.py │ ├── encoders │ │ ├── __init__.py │ │ ├── byte_bpe.py │ │ ├── byte_utils.py │ │ ├── bytes.py │ │ ├── characters.py │ │ ├── fastbpe.py │ │ ├── gpt2_bpe.py │ │ ├── gpt2_bpe_utils.py │ │ ├── hf_bert_bpe.py │ │ ├── hf_byte_bpe.py │ │ ├── moses_tokenizer.py │ │ ├── nltk_tokenizer.py │ │ ├── sentencepiece_bpe.py │ │ ├── space_tokenizer.py │ │ ├── subword_nmt_bpe.py │ │ └── utils.py │ ├── fairseq_dataset.py │ ├── id_dataset.py │ ├── indexed_dataset.py │ ├── iterators.py │ ├── language_pair_dataset.py │ ├── list_dataset.py │ ├── lm_context_window_dataset.py │ ├── lru_cache_dataset.py │ ├── mask_tokens_dataset.py │ ├── monolingual_dataset.py │ ├── multi_corpus_dataset.py │ ├── multi_corpus_sampled_dataset.py │ ├── nested_dictionary_dataset.py │ ├── num_samples_dataset.py │ ├── numel_dataset.py │ ├── offset_tokens_dataset.py │ ├── pad_dataset.py │ ├── plasma_utils.py │ ├── prepend_dataset.py │ ├── prepend_token_dataset.py │ ├── raw_label_dataset.py │ ├── replace_dataset.py │ ├── resampling_dataset.py │ ├── roll_dataset.py │ ├── round_robin_zip_datasets.py │ ├── shorten_dataset.py │ ├── sort_dataset.py │ ├── strip_token_dataset.py │ ├── subsample_dataset.py │ ├── t5_dataset.py │ ├── table_lookup_dataset.py │ ├── tensor_list_dataset.py │ ├── text_compressor.py │ ├── token_block_dataset.py │ ├── token_block_utils_fast.pyx │ ├── transform_eos_concat_langpair_dataset.py │ ├── transform_eos_dataset.py │ └── transform_eos_lang_pair_dataset.py ├── dataclass │ ├── __init__.py │ ├── configs.py │ ├── constants.py │ ├── initialize.py │ └── utils.py ├── distributed │ ├── __init__.py │ ├── distributed_timeout_wrapper.py │ ├── fully_sharded_data_parallel.py │ ├── legacy_distributed_data_parallel.py │ ├── module_proxy_wrapper.py │ ├── tpu_distributed_data_parallel.py │ └── utils.py ├── file_chunker_utils.py ├── file_io.py ├── file_utils.py ├── glue_utils.py ├── hub_utils.py ├── incremental_decoding_utils.py ├── iterative_refinement_generator.py ├── logging │ ├── __init__.py │ ├── meters.py │ ├── metrics.py │ └── progress_bar.py ├── model_parallel │ ├── __init__.py │ ├── criterions │ │ ├── __init__.py │ │ └── vocab_parallel_cross_entropy.py │ ├── megatron │ │ ├── .gitignore │ │ ├── LICENSE │ │ ├── README.md │ │ ├── arguments.py │ │ ├── configure_data.py │ │ ├── data_utils │ │ │ ├── __init__.py │ │ │ ├── corpora.py │ │ │ ├── datasets.py │ │ │ ├── file_utils.py │ │ │ ├── lazy_loader.py │ │ │ ├── samplers.py │ │ │ ├── tf_dl.py │ │ │ ├── tokenization.py │ │ │ ├── tokenization_gpt2.py │ │ │ └── wordpiece.py │ │ ├── detokenizer.py │ │ ├── docker │ │ │ ├── Dockerfile │ │ │ ├── README.md │ │ │ └── requirements.txt │ │ ├── evaluate_gpt2.py │ │ ├── fp16 │ │ │ ├── __init__.py │ │ │ ├── fp16.py │ │ │ ├── fp16util.py │ │ │ └── loss_scaler.py │ │ ├── generate_samples.py │ │ ├── gpt2_data_loader.py │ │ ├── learning_rates.py │ │ ├── model │ │ │ ├── __init__.py │ │ │ ├── distributed.py │ │ │ ├── gpt2_modeling.py │ │ │ ├── model.py │ │ │ └── modeling.py │ │ ├── mpu │ │ │ ├── __init__.py │ │ │ ├── cross_entropy.py │ │ │ ├── data.py │ │ │ ├── grads.py │ │ │ ├── initialize.py │ │ │ ├── layers.py │ │ │ ├── mappings.py │ │ │ ├── random.py │ │ │ ├── tests │ │ │ │ ├── __init__.py │ │ │ │ ├── commons.py │ │ │ │ ├── test_cross_entropy.py │ │ │ │ ├── test_data.py │ │ │ │ ├── test_initialize.py │ │ │ │ ├── test_layers.py │ │ │ │ └── test_random.py │ │ │ ├── transformer.py │ │ │ └── utils.py │ │ ├── openwebtext │ │ │ ├── README.md │ │ │ ├── blacklist_urls.py │ │ │ ├── cleanup_dataset.py │ │ │ ├── find_duplicates.py │ │ │ ├── group_duplicates_url.py │ │ │ ├── make_gpt2_dataset.py │ │ │ ├── make_gpt2_sizes.py │ │ │ ├── merge_jsons.py │ │ │ ├── remove_group_duplicates.py │ │ │ ├── run_make_gpt2_dataset.sh │ │ │ └── tokenizer.py │ │ ├── pretrain_bert.py │ │ ├── pretrain_gpt2.py │ │ ├── requirements.txt │ │ ├── scripts │ │ │ ├── generate_text.sh │ │ │ ├── presplit_sentences_json.py │ │ │ ├── pretrain_bert.sh │ │ │ ├── pretrain_bert_distributed.sh │ │ │ ├── pretrain_bert_model_parallel.sh │ │ │ ├── pretrain_bert_sentencepiece.sh │ │ │ ├── pretrain_bert_tfrecords_distributed.sh │ │ │ ├── pretrain_gpt2.sh │ │ │ ├── pretrain_gpt2_distributed.sh │ │ │ ├── pretrain_gpt2_model_parallel.sh │ │ │ ├── run_gpt2_eval.py │ │ │ ├── split_gpt2_json.py │ │ │ └── split_json.py │ │ └── utils.py │ ├── megatron_trainer.py │ ├── models │ │ ├── __init__.py │ │ ├── pipeline_parallel_transformer │ │ │ ├── __init__.py │ │ │ ├── layers.py │ │ │ └── model.py │ │ ├── roberta │ │ │ ├── __init__.py │ │ │ └── model.py │ │ ├── transformer.py │ │ └── transformer_lm.py │ └── modules │ │ ├── __init__.py │ │ ├── multihead_attention.py │ │ └── transformer_layer.py ├── models │ ├── __init__.py │ ├── composite_encoder.py │ ├── distributed_fairseq_model.py │ ├── ema │ │ ├── __init__.py │ │ └── ema.py │ ├── fairseq_decoder.py │ ├── fairseq_encoder.py │ ├── fairseq_incremental_decoder.py │ ├── fairseq_model.py │ ├── huggingface │ │ ├── __init__.py │ │ └── hf_gpt2.py │ ├── lstm.py │ ├── lstm_lm.py │ ├── masked_lm.py │ ├── model_utils.py │ ├── roberta │ │ ├── __init__.py │ │ ├── alignment_utils.py │ │ ├── enc_dec.py │ │ ├── hub_interface.py │ │ └── model.py │ ├── transformer │ │ ├── __init__.py │ │ ├── transformer_base.py │ │ ├── transformer_config.py │ │ ├── transformer_decoder.py │ │ ├── transformer_encoder.py │ │ ├── transformer_legacy.py │ │ └── transformer_t5.py │ └── transformer_lm.py ├── modules │ ├── __init__.py │ ├── beamable_mm.py │ ├── character_token_embedder.py │ ├── checkpoint_activations.py │ ├── cross_entropy.py │ ├── cuda_utils.cu │ ├── downsampled_multihead_attention.py │ ├── dynamic_crf_layer.py │ ├── espnet_multihead_attention.py │ ├── fairseq_dropout.py │ ├── fp32_batch_norm.py │ ├── fp32_group_norm.py │ ├── fp32_instance_norm.py │ ├── gelu.py │ ├── grad_multiply.py │ ├── kmeans_attention.py │ ├── layer_drop.py │ ├── layer_norm.py │ ├── learned_positional_embedding.py │ ├── lstm_cell_with_zoneout.py │ ├── multihead_attention.py │ ├── positional_embedding.py │ ├── positional_encoding.py │ ├── quant_noise.py │ ├── relative_positional_embedding.py │ ├── rotary_positional_embedding.py │ ├── same_pad.py │ ├── scalar_bias.py │ ├── sinusoidal_positional_embedding.py │ ├── transformer_layer.py │ ├── transformer_sentence_encoder.py │ ├── transformer_sentence_encoder_layer.py │ ├── transpose_last.py │ └── unfold.py ├── nan_detector.py ├── ngram_repeat_block.py ├── optim │ ├── __init__.py │ ├── adadelta.py │ ├── adafactor.py │ ├── adagrad.py │ ├── adam.py │ ├── adamax.py │ ├── amp_optimizer.py │ ├── bmuf.py │ ├── composite.py │ ├── cpu_adam.py │ ├── dynamic_loss_scaler.py │ ├── fairseq_optimizer.py │ ├── fp16_optimizer.py │ ├── fused_adam.py │ ├── fused_lamb.py │ ├── lr_scheduler │ │ ├── __init__.py │ │ ├── cosine_lr_scheduler.py │ │ ├── fairseq_lr_scheduler.py │ │ ├── fixed_schedule.py │ │ ├── inverse_square_root_schedule.py │ │ ├── manual_lr_scheduler.py │ │ ├── pass_through.py │ │ ├── polynomial_decay_schedule.py │ │ ├── reduce_lr_on_plateau.py │ │ ├── step_lr_scheduler.py │ │ ├── tri_stage_lr_scheduler.py │ │ └── triangular_lr_scheduler.py │ ├── nag.py │ ├── sgd.py │ └── shard.py ├── options.py ├── pdb.py ├── registry.py ├── scoring │ ├── __init__.py │ ├── bertscore.py │ ├── bleu.py │ ├── chrf.py │ ├── meteor.py │ ├── tokenizer.py │ └── wer.py ├── search.py ├── sequence_generator.py ├── sequence_scorer.py ├── tasks │ ├── __init__.py │ ├── fairseq_task.py │ ├── language_modeling.py │ ├── masked_lm.py │ ├── sentence_prediction.py │ ├── sentence_ranking.py │ ├── t5.py │ ├── t5_mixture.py │ ├── t5_sentence_prediction.py │ ├── t5_seq2seq.py │ ├── t5_seq2seq_mixture.py │ └── translation.py ├── token_generation_constraints.py ├── tokenizer.py ├── trainer.py ├── utils.py └── version.txt ├── fairseq_cli ├── __init__.py ├── eval_lm.py ├── generate.py ├── hydra_train.py ├── interactive.py ├── preprocess.py ├── score.py ├── train.py └── validate.py ├── fused_ops ├── csrc │ ├── fused_layer_norm │ │ ├── interface.cpp │ │ ├── interface_gamma_beta.cpp │ │ ├── layernorm.cu │ │ └── layernorm_backward.cu │ ├── fused_softmax_dropout │ │ ├── interface.cpp │ │ ├── softmax_dropout_kernel.cu │ │ └── softmax_fast.h │ └── type_shim.h └── setup.py ├── hubconf.py ├── pyproject.toml ├── scripts ├── __init__.py ├── average_checkpoints.py ├── compare_namespaces.py ├── compound_split_bleu.sh ├── constraints │ ├── extract.py │ └── validate.py ├── convert_dictionary.lua ├── convert_model.lua ├── count_docs.py ├── read_binarized.py ├── rm_pt.py ├── sacrebleu.sh ├── shard_docs.py ├── split_train_valid_docs.py ├── spm_decode.py ├── spm_encode.py ├── spm_train.py └── test_fsdp.sh ├── setup.cfg ├── setup.py ├── tests ├── __init__.py ├── distributed │ ├── __init__.py │ ├── test_bmuf.py │ ├── test_distributed_timeout_wrapper.py │ ├── test_module_proxy_wrapper.py │ ├── test_utils.py │ └── utils.py ├── gpu │ ├── __init__.py │ ├── test_binaries_gpu.py │ ├── test_ema_gpu.py │ └── transformer_quantization_config.yaml ├── test_activation_checkpointing.py ├── test_amp_optimizer.py ├── test_average_checkpoints.py ├── test_backtranslation_dataset.py ├── test_binaries.py ├── test_binarizer.py ├── test_character_token_embedder.py ├── test_checkpoint_utils.py ├── test_concat_dataset.py ├── test_constraints.py ├── test_convtbc.py ├── test_data_utils.py ├── test_dataclass_utils.py ├── test_dataset.py ├── test_dictionary.py ├── test_ema.py ├── test_espnet_multihead_attention.py ├── test_export.py ├── test_file_chunker_utils.py ├── test_file_io.py ├── test_fp16_optimizer.py ├── test_inference_dropout.py ├── test_iopath.py ├── test_iterators.py ├── test_label_smoothing.py ├── test_lm_context_window.py ├── test_lstm_jitable.py ├── test_memory_efficient_fp16.py ├── test_metrics.py ├── test_multi_corpus_dataset.py ├── test_multi_corpus_sampled_dataset.py ├── test_multihead_attention.py ├── test_noising.py ├── test_online_backtranslation.py ├── test_plasma_utils.py ├── test_positional_encoding.py ├── test_reproducibility.py ├── test_resampling_dataset.py ├── test_roberta.py ├── test_rotary_positional_embedding.py ├── test_sequence_generator.py ├── test_sequence_scorer.py ├── test_sparse_multihead_attention.py ├── test_token_block_dataset.py ├── test_train.py ├── test_transformer.py ├── test_utils.py ├── test_valid_subset_checks.py └── utils.py └── train.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/.gitignore -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/README.md -------------------------------------------------------------------------------- /eval/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/README.md -------------------------------------------------------------------------------- /eval/evaluate_humaneval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/evaluate_humaneval.py -------------------------------------------------------------------------------- /eval/evaluate_humaneval_plus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/evaluate_humaneval_plus.py -------------------------------------------------------------------------------- /eval/evaluate_mbpp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/evaluate_mbpp.py -------------------------------------------------------------------------------- /eval/evaluate_mbpp_plus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/evaluate_mbpp_plus.py -------------------------------------------------------------------------------- /eval/run_humaneval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/run_humaneval.py -------------------------------------------------------------------------------- /eval/run_humaneval_plus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/run_humaneval_plus.py -------------------------------------------------------------------------------- /eval/run_mbpp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/run_mbpp.py -------------------------------------------------------------------------------- /eval/run_mbpp_plus.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/eval/run_mbpp_plus.py -------------------------------------------------------------------------------- /model_release/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/LICENSE -------------------------------------------------------------------------------- /model_release/pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/pyproject.toml -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/configuration_fairseq_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/src/fairseq_t5/configuration_fairseq_t5.py -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/convert_to_hf_checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/src/fairseq_t5/convert_to_hf_checkpoint.py -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/fairseq_dictionary.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/src/fairseq_t5/fairseq_dictionary.py -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/modeling_fairseq_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/src/fairseq_t5/modeling_fairseq_t5.py -------------------------------------------------------------------------------- /model_release/src/fairseq_t5/tokenization_ast_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/model_release/src/fairseq_t5/tokenization_ast_t5.py -------------------------------------------------------------------------------- /training/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/README.md -------------------------------------------------------------------------------- /training/examples/.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/examples/.gitignore -------------------------------------------------------------------------------- /training/examples/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/examples/__init__.py -------------------------------------------------------------------------------- /training/examples/t5/config/pl/base1m_rpe_10_2e-4_64gpus_bs8192.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/examples/t5/config/pl/base1m_rpe_10_2e-4_64gpus_bs8192.yaml -------------------------------------------------------------------------------- /training/examples/t5/config/pl/base1m_rpe_mask25_ast10_lb5_ub100_op50_or50_2e-4_64gpus_bs8192.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/examples/t5/config/pl/base1m_rpe_mask25_ast10_lb5_ub100_op50_or50_2e-4_64gpus_bs8192.yaml -------------------------------------------------------------------------------- /training/fairseq/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/__init__.py -------------------------------------------------------------------------------- /training/fairseq/ast_t5_dataset_fast.pyi: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/ast_t5_dataset_fast.pyi -------------------------------------------------------------------------------- /training/fairseq/benchmark/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/__init__.py -------------------------------------------------------------------------------- /training/fairseq/benchmark/dummy_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/dummy_dataset.py -------------------------------------------------------------------------------- /training/fairseq/benchmark/dummy_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/dummy_lm.py -------------------------------------------------------------------------------- /training/fairseq/benchmark/dummy_masked_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/dummy_masked_lm.py -------------------------------------------------------------------------------- /training/fairseq/benchmark/dummy_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/dummy_model.py -------------------------------------------------------------------------------- /training/fairseq/benchmark/dummy_mt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/benchmark/dummy_mt.py -------------------------------------------------------------------------------- /training/fairseq/binarizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/binarizer.py -------------------------------------------------------------------------------- /training/fairseq/checkpoint_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/checkpoint_utils.py -------------------------------------------------------------------------------- /training/fairseq/clib/cuda/ngram_repeat_block_cuda.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/cuda/ngram_repeat_block_cuda.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/cuda/ngram_repeat_block_cuda_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/cuda/ngram_repeat_block_cuda_kernel.cu -------------------------------------------------------------------------------- /training/fairseq/clib/libast/interface.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libast/interface.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libast/libast.hpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libast/libast.hpp -------------------------------------------------------------------------------- /training/fairseq/clib/libbase/balanced_assignment.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libbase/balanced_assignment.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libbleu/libbleu.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libbleu/libbleu.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libbleu/module.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libbleu/module.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libnat/edit_dist.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libnat/edit_dist.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libnat_cuda/binding.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libnat_cuda/binding.cpp -------------------------------------------------------------------------------- /training/fairseq/clib/libnat_cuda/edit_dist.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libnat_cuda/edit_dist.cu -------------------------------------------------------------------------------- /training/fairseq/clib/libnat_cuda/edit_dist.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/clib/libnat_cuda/edit_dist.h -------------------------------------------------------------------------------- /training/fairseq/config/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/__init__.py -------------------------------------------------------------------------------- /training/fairseq/config/config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/config.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_big.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_big.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_gbw.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_gbw.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_gpt.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_gpt.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_big.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_big.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_medium.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_medium.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_small.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_gpt2_small.yaml -------------------------------------------------------------------------------- /training/fairseq/config/model/transformer_lm/transformer_lm_wiki103.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/config/model/transformer_lm/transformer_lm_wiki103.yaml -------------------------------------------------------------------------------- /training/fairseq/criterions/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/__init__.py -------------------------------------------------------------------------------- /training/fairseq/criterions/fairseq_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/fairseq_criterion.py -------------------------------------------------------------------------------- /training/fairseq/criterions/label_smoothed_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/label_smoothed_cross_entropy.py -------------------------------------------------------------------------------- /training/fairseq/criterions/masked_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/masked_lm.py -------------------------------------------------------------------------------- /training/fairseq/criterions/model_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/model_criterion.py -------------------------------------------------------------------------------- /training/fairseq/criterions/sentence_prediction.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/sentence_prediction.py -------------------------------------------------------------------------------- /training/fairseq/criterions/sentence_ranking.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/criterions/sentence_ranking.py -------------------------------------------------------------------------------- /training/fairseq/data/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/__init__.py -------------------------------------------------------------------------------- /training/fairseq/data/add_target_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/add_target_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/append_token_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/append_token_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/ast_t5_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/ast_t5_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/base_wrapper_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/base_wrapper_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/bucket_pad_length_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/bucket_pad_length_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/colorize_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/colorize_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/concat_capped_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/concat_capped_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/concat_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/concat_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/concat_sentences_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/concat_sentences_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/concat_weighted_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/concat_weighted_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/constant_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/constant_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/data_utils.py -------------------------------------------------------------------------------- /training/fairseq/data/data_utils_fast.pyx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/data_utils_fast.pyx -------------------------------------------------------------------------------- /training/fairseq/data/dictionary.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/dictionary.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/__init__.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/byte_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/byte_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/byte_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/byte_utils.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/bytes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/bytes.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/characters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/characters.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/fastbpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/fastbpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/gpt2_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/gpt2_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/gpt2_bpe_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/gpt2_bpe_utils.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/hf_bert_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/hf_bert_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/hf_byte_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/hf_byte_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/moses_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/moses_tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/nltk_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/nltk_tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/sentencepiece_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/sentencepiece_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/space_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/space_tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/subword_nmt_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/subword_nmt_bpe.py -------------------------------------------------------------------------------- /training/fairseq/data/encoders/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/encoders/utils.py -------------------------------------------------------------------------------- /training/fairseq/data/fairseq_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/fairseq_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/id_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/id_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/indexed_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/indexed_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/iterators.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/iterators.py -------------------------------------------------------------------------------- /training/fairseq/data/language_pair_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/language_pair_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/list_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/list_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/lm_context_window_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/lm_context_window_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/lru_cache_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/lru_cache_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/mask_tokens_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/mask_tokens_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/monolingual_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/monolingual_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/multi_corpus_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/multi_corpus_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/multi_corpus_sampled_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/multi_corpus_sampled_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/nested_dictionary_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/nested_dictionary_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/num_samples_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/num_samples_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/numel_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/numel_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/offset_tokens_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/offset_tokens_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/pad_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/pad_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/plasma_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/plasma_utils.py -------------------------------------------------------------------------------- /training/fairseq/data/prepend_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/prepend_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/prepend_token_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/prepend_token_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/raw_label_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/raw_label_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/replace_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/replace_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/resampling_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/resampling_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/roll_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/roll_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/round_robin_zip_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/round_robin_zip_datasets.py -------------------------------------------------------------------------------- /training/fairseq/data/shorten_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/shorten_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/sort_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/sort_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/strip_token_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/strip_token_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/subsample_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/subsample_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/t5_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/t5_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/table_lookup_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/table_lookup_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/tensor_list_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/tensor_list_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/text_compressor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/text_compressor.py -------------------------------------------------------------------------------- /training/fairseq/data/token_block_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/token_block_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/token_block_utils_fast.pyx: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/token_block_utils_fast.pyx -------------------------------------------------------------------------------- /training/fairseq/data/transform_eos_concat_langpair_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/transform_eos_concat_langpair_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/transform_eos_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/transform_eos_dataset.py -------------------------------------------------------------------------------- /training/fairseq/data/transform_eos_lang_pair_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/data/transform_eos_lang_pair_dataset.py -------------------------------------------------------------------------------- /training/fairseq/dataclass/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/dataclass/__init__.py -------------------------------------------------------------------------------- /training/fairseq/dataclass/configs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/dataclass/configs.py -------------------------------------------------------------------------------- /training/fairseq/dataclass/constants.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/dataclass/constants.py -------------------------------------------------------------------------------- /training/fairseq/dataclass/initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/dataclass/initialize.py -------------------------------------------------------------------------------- /training/fairseq/dataclass/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/dataclass/utils.py -------------------------------------------------------------------------------- /training/fairseq/distributed/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/__init__.py -------------------------------------------------------------------------------- /training/fairseq/distributed/distributed_timeout_wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/distributed_timeout_wrapper.py -------------------------------------------------------------------------------- /training/fairseq/distributed/fully_sharded_data_parallel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/fully_sharded_data_parallel.py -------------------------------------------------------------------------------- /training/fairseq/distributed/legacy_distributed_data_parallel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/legacy_distributed_data_parallel.py -------------------------------------------------------------------------------- /training/fairseq/distributed/module_proxy_wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/module_proxy_wrapper.py -------------------------------------------------------------------------------- /training/fairseq/distributed/tpu_distributed_data_parallel.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/tpu_distributed_data_parallel.py -------------------------------------------------------------------------------- /training/fairseq/distributed/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/distributed/utils.py -------------------------------------------------------------------------------- /training/fairseq/file_chunker_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/file_chunker_utils.py -------------------------------------------------------------------------------- /training/fairseq/file_io.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/file_io.py -------------------------------------------------------------------------------- /training/fairseq/file_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/file_utils.py -------------------------------------------------------------------------------- /training/fairseq/glue_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/glue_utils.py -------------------------------------------------------------------------------- /training/fairseq/hub_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/hub_utils.py -------------------------------------------------------------------------------- /training/fairseq/incremental_decoding_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/incremental_decoding_utils.py -------------------------------------------------------------------------------- /training/fairseq/iterative_refinement_generator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/iterative_refinement_generator.py -------------------------------------------------------------------------------- /training/fairseq/logging/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/fairseq/logging/meters.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/logging/meters.py -------------------------------------------------------------------------------- /training/fairseq/logging/metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/logging/metrics.py -------------------------------------------------------------------------------- /training/fairseq/logging/progress_bar.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/logging/progress_bar.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/criterions/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/criterions/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/criterions/vocab_parallel_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/criterions/vocab_parallel_cross_entropy.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/.gitignore: -------------------------------------------------------------------------------- 1 | __pycache__ 2 | -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/LICENSE -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/README.md -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/arguments.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/configure_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/configure_data.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/corpora.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/corpora.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/datasets.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/file_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/file_utils.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/lazy_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/lazy_loader.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/samplers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/samplers.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/tf_dl.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/tf_dl.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/tokenization.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/tokenization_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/tokenization_gpt2.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/data_utils/wordpiece.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/data_utils/wordpiece.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/detokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/detokenizer.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/docker/Dockerfile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/docker/Dockerfile -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/docker/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/docker/README.md -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/docker/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/docker/requirements.txt -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/evaluate_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/evaluate_gpt2.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/fp16/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/fp16/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/fp16/fp16.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/fp16/fp16.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/fp16/fp16util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/fp16/fp16util.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/fp16/loss_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/fp16/loss_scaler.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/generate_samples.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/generate_samples.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/gpt2_data_loader.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/gpt2_data_loader.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/learning_rates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/learning_rates.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/model/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/model/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/model/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/model/distributed.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/model/gpt2_modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/model/gpt2_modeling.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/model/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/model/model.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/model/modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/model/modeling.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/cross_entropy.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/data.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/grads.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/grads.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/initialize.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/layers.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/mappings.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/random.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/commons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/commons.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/test_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/test_cross_entropy.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/test_data.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/test_initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/test_initialize.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/test_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/test_layers.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/tests/test_random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/tests/test_random.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/transformer.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/mpu/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/mpu/utils.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/README.md -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/blacklist_urls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/blacklist_urls.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/cleanup_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/cleanup_dataset.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/find_duplicates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/find_duplicates.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/group_duplicates_url.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/group_duplicates_url.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/make_gpt2_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/make_gpt2_dataset.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/make_gpt2_sizes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/make_gpt2_sizes.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/merge_jsons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/merge_jsons.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/remove_group_duplicates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/remove_group_duplicates.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/run_make_gpt2_dataset.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/run_make_gpt2_dataset.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/openwebtext/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/openwebtext/tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/pretrain_bert.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/pretrain_bert.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/pretrain_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/pretrain_gpt2.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/requirements.txt -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/generate_text.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/generate_text.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/presplit_sentences_json.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/presplit_sentences_json.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_bert.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_bert.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_bert_distributed.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_bert_distributed.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_bert_model_parallel.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_bert_model_parallel.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_bert_sentencepiece.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_bert_sentencepiece.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_bert_tfrecords_distributed.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_bert_tfrecords_distributed.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2_distributed.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2_distributed.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2_model_parallel.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/pretrain_gpt2_model_parallel.sh -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/run_gpt2_eval.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/run_gpt2_eval.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/split_gpt2_json.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/split_gpt2_json.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/scripts/split_json.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/scripts/split_json.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron/utils.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/megatron_trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/megatron_trainer.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/pipeline_parallel_transformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/pipeline_parallel_transformer/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/pipeline_parallel_transformer/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/pipeline_parallel_transformer/layers.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/pipeline_parallel_transformer/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/pipeline_parallel_transformer/model.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/roberta/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/roberta/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/roberta/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/roberta/model.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/transformer.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/models/transformer_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/models/transformer_lm.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/modules/__init__.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/modules/multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/modules/multihead_attention.py -------------------------------------------------------------------------------- /training/fairseq/model_parallel/modules/transformer_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/model_parallel/modules/transformer_layer.py -------------------------------------------------------------------------------- /training/fairseq/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/__init__.py -------------------------------------------------------------------------------- /training/fairseq/models/composite_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/composite_encoder.py -------------------------------------------------------------------------------- /training/fairseq/models/distributed_fairseq_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/distributed_fairseq_model.py -------------------------------------------------------------------------------- /training/fairseq/models/ema/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/ema/__init__.py -------------------------------------------------------------------------------- /training/fairseq/models/ema/ema.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/ema/ema.py -------------------------------------------------------------------------------- /training/fairseq/models/fairseq_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/fairseq_decoder.py -------------------------------------------------------------------------------- /training/fairseq/models/fairseq_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/fairseq_encoder.py -------------------------------------------------------------------------------- /training/fairseq/models/fairseq_incremental_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/fairseq_incremental_decoder.py -------------------------------------------------------------------------------- /training/fairseq/models/fairseq_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/fairseq_model.py -------------------------------------------------------------------------------- /training/fairseq/models/huggingface/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/huggingface/__init__.py -------------------------------------------------------------------------------- /training/fairseq/models/huggingface/hf_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/huggingface/hf_gpt2.py -------------------------------------------------------------------------------- /training/fairseq/models/lstm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/lstm.py -------------------------------------------------------------------------------- /training/fairseq/models/lstm_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/lstm_lm.py -------------------------------------------------------------------------------- /training/fairseq/models/masked_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/masked_lm.py -------------------------------------------------------------------------------- /training/fairseq/models/model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/model_utils.py -------------------------------------------------------------------------------- /training/fairseq/models/roberta/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/roberta/__init__.py -------------------------------------------------------------------------------- /training/fairseq/models/roberta/alignment_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/roberta/alignment_utils.py -------------------------------------------------------------------------------- /training/fairseq/models/roberta/enc_dec.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/roberta/enc_dec.py -------------------------------------------------------------------------------- /training/fairseq/models/roberta/hub_interface.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/roberta/hub_interface.py -------------------------------------------------------------------------------- /training/fairseq/models/roberta/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/roberta/model.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/__init__.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_base.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_config.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_decoder.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_encoder.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_legacy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_legacy.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer/transformer_t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer/transformer_t5.py -------------------------------------------------------------------------------- /training/fairseq/models/transformer_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/models/transformer_lm.py -------------------------------------------------------------------------------- /training/fairseq/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/__init__.py -------------------------------------------------------------------------------- /training/fairseq/modules/beamable_mm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/beamable_mm.py -------------------------------------------------------------------------------- /training/fairseq/modules/character_token_embedder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/character_token_embedder.py -------------------------------------------------------------------------------- /training/fairseq/modules/checkpoint_activations.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/checkpoint_activations.py -------------------------------------------------------------------------------- /training/fairseq/modules/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/cross_entropy.py -------------------------------------------------------------------------------- /training/fairseq/modules/cuda_utils.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/cuda_utils.cu -------------------------------------------------------------------------------- /training/fairseq/modules/downsampled_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/downsampled_multihead_attention.py -------------------------------------------------------------------------------- /training/fairseq/modules/dynamic_crf_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/dynamic_crf_layer.py -------------------------------------------------------------------------------- /training/fairseq/modules/espnet_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/espnet_multihead_attention.py -------------------------------------------------------------------------------- /training/fairseq/modules/fairseq_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/fairseq_dropout.py -------------------------------------------------------------------------------- /training/fairseq/modules/fp32_batch_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/fp32_batch_norm.py -------------------------------------------------------------------------------- /training/fairseq/modules/fp32_group_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/fp32_group_norm.py -------------------------------------------------------------------------------- /training/fairseq/modules/fp32_instance_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/fp32_instance_norm.py -------------------------------------------------------------------------------- /training/fairseq/modules/gelu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/gelu.py -------------------------------------------------------------------------------- /training/fairseq/modules/grad_multiply.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/grad_multiply.py -------------------------------------------------------------------------------- /training/fairseq/modules/kmeans_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/kmeans_attention.py -------------------------------------------------------------------------------- /training/fairseq/modules/layer_drop.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/layer_drop.py -------------------------------------------------------------------------------- /training/fairseq/modules/layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/layer_norm.py -------------------------------------------------------------------------------- /training/fairseq/modules/learned_positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/learned_positional_embedding.py -------------------------------------------------------------------------------- /training/fairseq/modules/lstm_cell_with_zoneout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/lstm_cell_with_zoneout.py -------------------------------------------------------------------------------- /training/fairseq/modules/multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/multihead_attention.py -------------------------------------------------------------------------------- /training/fairseq/modules/positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/positional_embedding.py -------------------------------------------------------------------------------- /training/fairseq/modules/positional_encoding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/positional_encoding.py -------------------------------------------------------------------------------- /training/fairseq/modules/quant_noise.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/quant_noise.py -------------------------------------------------------------------------------- /training/fairseq/modules/relative_positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/relative_positional_embedding.py -------------------------------------------------------------------------------- /training/fairseq/modules/rotary_positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/rotary_positional_embedding.py -------------------------------------------------------------------------------- /training/fairseq/modules/same_pad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/same_pad.py -------------------------------------------------------------------------------- /training/fairseq/modules/scalar_bias.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/scalar_bias.py -------------------------------------------------------------------------------- /training/fairseq/modules/sinusoidal_positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/sinusoidal_positional_embedding.py -------------------------------------------------------------------------------- /training/fairseq/modules/transformer_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/transformer_layer.py -------------------------------------------------------------------------------- /training/fairseq/modules/transformer_sentence_encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/transformer_sentence_encoder.py -------------------------------------------------------------------------------- /training/fairseq/modules/transformer_sentence_encoder_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/transformer_sentence_encoder_layer.py -------------------------------------------------------------------------------- /training/fairseq/modules/transpose_last.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/transpose_last.py -------------------------------------------------------------------------------- /training/fairseq/modules/unfold.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/modules/unfold.py -------------------------------------------------------------------------------- /training/fairseq/nan_detector.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/nan_detector.py -------------------------------------------------------------------------------- /training/fairseq/ngram_repeat_block.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/ngram_repeat_block.py -------------------------------------------------------------------------------- /training/fairseq/optim/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/__init__.py -------------------------------------------------------------------------------- /training/fairseq/optim/adadelta.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/adadelta.py -------------------------------------------------------------------------------- /training/fairseq/optim/adafactor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/adafactor.py -------------------------------------------------------------------------------- /training/fairseq/optim/adagrad.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/adagrad.py -------------------------------------------------------------------------------- /training/fairseq/optim/adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/adam.py -------------------------------------------------------------------------------- /training/fairseq/optim/adamax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/adamax.py -------------------------------------------------------------------------------- /training/fairseq/optim/amp_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/amp_optimizer.py -------------------------------------------------------------------------------- /training/fairseq/optim/bmuf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/bmuf.py -------------------------------------------------------------------------------- /training/fairseq/optim/composite.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/composite.py -------------------------------------------------------------------------------- /training/fairseq/optim/cpu_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/cpu_adam.py -------------------------------------------------------------------------------- /training/fairseq/optim/dynamic_loss_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/dynamic_loss_scaler.py -------------------------------------------------------------------------------- /training/fairseq/optim/fairseq_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/fairseq_optimizer.py -------------------------------------------------------------------------------- /training/fairseq/optim/fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/fp16_optimizer.py -------------------------------------------------------------------------------- /training/fairseq/optim/fused_adam.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/fused_adam.py -------------------------------------------------------------------------------- /training/fairseq/optim/fused_lamb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/fused_lamb.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/__init__.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/cosine_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/cosine_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/fairseq_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/fairseq_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/fixed_schedule.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/fixed_schedule.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/inverse_square_root_schedule.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/inverse_square_root_schedule.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/manual_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/manual_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/pass_through.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/pass_through.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/polynomial_decay_schedule.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/polynomial_decay_schedule.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/reduce_lr_on_plateau.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/reduce_lr_on_plateau.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/step_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/step_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/tri_stage_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/tri_stage_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/lr_scheduler/triangular_lr_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/lr_scheduler/triangular_lr_scheduler.py -------------------------------------------------------------------------------- /training/fairseq/optim/nag.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/nag.py -------------------------------------------------------------------------------- /training/fairseq/optim/sgd.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/sgd.py -------------------------------------------------------------------------------- /training/fairseq/optim/shard.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/optim/shard.py -------------------------------------------------------------------------------- /training/fairseq/options.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/options.py -------------------------------------------------------------------------------- /training/fairseq/pdb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/pdb.py -------------------------------------------------------------------------------- /training/fairseq/registry.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/registry.py -------------------------------------------------------------------------------- /training/fairseq/scoring/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/__init__.py -------------------------------------------------------------------------------- /training/fairseq/scoring/bertscore.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/bertscore.py -------------------------------------------------------------------------------- /training/fairseq/scoring/bleu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/bleu.py -------------------------------------------------------------------------------- /training/fairseq/scoring/chrf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/chrf.py -------------------------------------------------------------------------------- /training/fairseq/scoring/meteor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/meteor.py -------------------------------------------------------------------------------- /training/fairseq/scoring/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/scoring/wer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/scoring/wer.py -------------------------------------------------------------------------------- /training/fairseq/search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/search.py -------------------------------------------------------------------------------- /training/fairseq/sequence_generator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/sequence_generator.py -------------------------------------------------------------------------------- /training/fairseq/sequence_scorer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/sequence_scorer.py -------------------------------------------------------------------------------- /training/fairseq/tasks/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/__init__.py -------------------------------------------------------------------------------- /training/fairseq/tasks/fairseq_task.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/fairseq_task.py -------------------------------------------------------------------------------- /training/fairseq/tasks/language_modeling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/language_modeling.py -------------------------------------------------------------------------------- /training/fairseq/tasks/masked_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/masked_lm.py -------------------------------------------------------------------------------- /training/fairseq/tasks/sentence_prediction.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/sentence_prediction.py -------------------------------------------------------------------------------- /training/fairseq/tasks/sentence_ranking.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/sentence_ranking.py -------------------------------------------------------------------------------- /training/fairseq/tasks/t5.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/t5.py -------------------------------------------------------------------------------- /training/fairseq/tasks/t5_mixture.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/t5_mixture.py -------------------------------------------------------------------------------- /training/fairseq/tasks/t5_sentence_prediction.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/t5_sentence_prediction.py -------------------------------------------------------------------------------- /training/fairseq/tasks/t5_seq2seq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/t5_seq2seq.py -------------------------------------------------------------------------------- /training/fairseq/tasks/t5_seq2seq_mixture.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/t5_seq2seq_mixture.py -------------------------------------------------------------------------------- /training/fairseq/tasks/translation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tasks/translation.py -------------------------------------------------------------------------------- /training/fairseq/token_generation_constraints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/token_generation_constraints.py -------------------------------------------------------------------------------- /training/fairseq/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/tokenizer.py -------------------------------------------------------------------------------- /training/fairseq/trainer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/trainer.py -------------------------------------------------------------------------------- /training/fairseq/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq/utils.py -------------------------------------------------------------------------------- /training/fairseq/version.txt: -------------------------------------------------------------------------------- 1 | 1.0.0a0 2 | -------------------------------------------------------------------------------- /training/fairseq_cli/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/fairseq_cli/eval_lm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/eval_lm.py -------------------------------------------------------------------------------- /training/fairseq_cli/generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/generate.py -------------------------------------------------------------------------------- /training/fairseq_cli/hydra_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/hydra_train.py -------------------------------------------------------------------------------- /training/fairseq_cli/interactive.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/interactive.py -------------------------------------------------------------------------------- /training/fairseq_cli/preprocess.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/preprocess.py -------------------------------------------------------------------------------- /training/fairseq_cli/score.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/score.py -------------------------------------------------------------------------------- /training/fairseq_cli/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/train.py -------------------------------------------------------------------------------- /training/fairseq_cli/validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fairseq_cli/validate.py -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_layer_norm/interface.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_layer_norm/interface.cpp -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_layer_norm/interface_gamma_beta.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_layer_norm/interface_gamma_beta.cpp -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_layer_norm/layernorm.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_layer_norm/layernorm.cu -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_layer_norm/layernorm_backward.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_layer_norm/layernorm_backward.cu -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_softmax_dropout/interface.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_softmax_dropout/interface.cpp -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_softmax_dropout/softmax_dropout_kernel.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_softmax_dropout/softmax_dropout_kernel.cu -------------------------------------------------------------------------------- /training/fused_ops/csrc/fused_softmax_dropout/softmax_fast.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/fused_softmax_dropout/softmax_fast.h -------------------------------------------------------------------------------- /training/fused_ops/csrc/type_shim.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/csrc/type_shim.h -------------------------------------------------------------------------------- /training/fused_ops/setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/fused_ops/setup.py -------------------------------------------------------------------------------- /training/hubconf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/hubconf.py -------------------------------------------------------------------------------- /training/pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/pyproject.toml -------------------------------------------------------------------------------- /training/scripts/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/scripts/average_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/average_checkpoints.py -------------------------------------------------------------------------------- /training/scripts/compare_namespaces.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/compare_namespaces.py -------------------------------------------------------------------------------- /training/scripts/compound_split_bleu.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/compound_split_bleu.sh -------------------------------------------------------------------------------- /training/scripts/constraints/extract.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/constraints/extract.py -------------------------------------------------------------------------------- /training/scripts/constraints/validate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/constraints/validate.py -------------------------------------------------------------------------------- /training/scripts/convert_dictionary.lua: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/convert_dictionary.lua -------------------------------------------------------------------------------- /training/scripts/convert_model.lua: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/convert_model.lua -------------------------------------------------------------------------------- /training/scripts/count_docs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/count_docs.py -------------------------------------------------------------------------------- /training/scripts/read_binarized.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/read_binarized.py -------------------------------------------------------------------------------- /training/scripts/rm_pt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/rm_pt.py -------------------------------------------------------------------------------- /training/scripts/sacrebleu.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/sacrebleu.sh -------------------------------------------------------------------------------- /training/scripts/shard_docs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/shard_docs.py -------------------------------------------------------------------------------- /training/scripts/split_train_valid_docs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/split_train_valid_docs.py -------------------------------------------------------------------------------- /training/scripts/spm_decode.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/spm_decode.py -------------------------------------------------------------------------------- /training/scripts/spm_encode.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/spm_encode.py -------------------------------------------------------------------------------- /training/scripts/spm_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/spm_train.py -------------------------------------------------------------------------------- /training/scripts/test_fsdp.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/scripts/test_fsdp.sh -------------------------------------------------------------------------------- /training/setup.cfg: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/setup.cfg -------------------------------------------------------------------------------- /training/setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/setup.py -------------------------------------------------------------------------------- /training/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/tests/distributed/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/tests/distributed/test_bmuf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/distributed/test_bmuf.py -------------------------------------------------------------------------------- /training/tests/distributed/test_distributed_timeout_wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/distributed/test_distributed_timeout_wrapper.py -------------------------------------------------------------------------------- /training/tests/distributed/test_module_proxy_wrapper.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/distributed/test_module_proxy_wrapper.py -------------------------------------------------------------------------------- /training/tests/distributed/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/distributed/test_utils.py -------------------------------------------------------------------------------- /training/tests/distributed/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/distributed/utils.py -------------------------------------------------------------------------------- /training/tests/gpu/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /training/tests/gpu/test_binaries_gpu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/gpu/test_binaries_gpu.py -------------------------------------------------------------------------------- /training/tests/gpu/test_ema_gpu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/gpu/test_ema_gpu.py -------------------------------------------------------------------------------- /training/tests/gpu/transformer_quantization_config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/gpu/transformer_quantization_config.yaml -------------------------------------------------------------------------------- /training/tests/test_activation_checkpointing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_activation_checkpointing.py -------------------------------------------------------------------------------- /training/tests/test_amp_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_amp_optimizer.py -------------------------------------------------------------------------------- /training/tests/test_average_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_average_checkpoints.py -------------------------------------------------------------------------------- /training/tests/test_backtranslation_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_backtranslation_dataset.py -------------------------------------------------------------------------------- /training/tests/test_binaries.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_binaries.py -------------------------------------------------------------------------------- /training/tests/test_binarizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_binarizer.py -------------------------------------------------------------------------------- /training/tests/test_character_token_embedder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_character_token_embedder.py -------------------------------------------------------------------------------- /training/tests/test_checkpoint_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_checkpoint_utils.py -------------------------------------------------------------------------------- /training/tests/test_concat_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_concat_dataset.py -------------------------------------------------------------------------------- /training/tests/test_constraints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_constraints.py -------------------------------------------------------------------------------- /training/tests/test_convtbc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_convtbc.py -------------------------------------------------------------------------------- /training/tests/test_data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_data_utils.py -------------------------------------------------------------------------------- /training/tests/test_dataclass_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_dataclass_utils.py -------------------------------------------------------------------------------- /training/tests/test_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_dataset.py -------------------------------------------------------------------------------- /training/tests/test_dictionary.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_dictionary.py -------------------------------------------------------------------------------- /training/tests/test_ema.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_ema.py -------------------------------------------------------------------------------- /training/tests/test_espnet_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_espnet_multihead_attention.py -------------------------------------------------------------------------------- /training/tests/test_export.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_export.py -------------------------------------------------------------------------------- /training/tests/test_file_chunker_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_file_chunker_utils.py -------------------------------------------------------------------------------- /training/tests/test_file_io.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_file_io.py -------------------------------------------------------------------------------- /training/tests/test_fp16_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_fp16_optimizer.py -------------------------------------------------------------------------------- /training/tests/test_inference_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_inference_dropout.py -------------------------------------------------------------------------------- /training/tests/test_iopath.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_iopath.py -------------------------------------------------------------------------------- /training/tests/test_iterators.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_iterators.py -------------------------------------------------------------------------------- /training/tests/test_label_smoothing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_label_smoothing.py -------------------------------------------------------------------------------- /training/tests/test_lm_context_window.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_lm_context_window.py -------------------------------------------------------------------------------- /training/tests/test_lstm_jitable.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_lstm_jitable.py -------------------------------------------------------------------------------- /training/tests/test_memory_efficient_fp16.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_memory_efficient_fp16.py -------------------------------------------------------------------------------- /training/tests/test_metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_metrics.py -------------------------------------------------------------------------------- /training/tests/test_multi_corpus_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_multi_corpus_dataset.py -------------------------------------------------------------------------------- /training/tests/test_multi_corpus_sampled_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_multi_corpus_sampled_dataset.py -------------------------------------------------------------------------------- /training/tests/test_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_multihead_attention.py -------------------------------------------------------------------------------- /training/tests/test_noising.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_noising.py -------------------------------------------------------------------------------- /training/tests/test_online_backtranslation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_online_backtranslation.py -------------------------------------------------------------------------------- /training/tests/test_plasma_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_plasma_utils.py -------------------------------------------------------------------------------- /training/tests/test_positional_encoding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_positional_encoding.py -------------------------------------------------------------------------------- /training/tests/test_reproducibility.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_reproducibility.py -------------------------------------------------------------------------------- /training/tests/test_resampling_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_resampling_dataset.py -------------------------------------------------------------------------------- /training/tests/test_roberta.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_roberta.py -------------------------------------------------------------------------------- /training/tests/test_rotary_positional_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_rotary_positional_embedding.py -------------------------------------------------------------------------------- /training/tests/test_sequence_generator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_sequence_generator.py -------------------------------------------------------------------------------- /training/tests/test_sequence_scorer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_sequence_scorer.py -------------------------------------------------------------------------------- /training/tests/test_sparse_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_sparse_multihead_attention.py -------------------------------------------------------------------------------- /training/tests/test_token_block_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_token_block_dataset.py -------------------------------------------------------------------------------- /training/tests/test_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_train.py -------------------------------------------------------------------------------- /training/tests/test_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_transformer.py -------------------------------------------------------------------------------- /training/tests/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_utils.py -------------------------------------------------------------------------------- /training/tests/test_valid_subset_checks.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/test_valid_subset_checks.py -------------------------------------------------------------------------------- /training/tests/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/tests/utils.py -------------------------------------------------------------------------------- /training/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/gonglinyuan/ast_t5/HEAD/training/train.py --------------------------------------------------------------------------------