├── .coveragerc ├── .github └── workflows │ └── python.yml ├── .gitignore ├── .gitlab-ci.yml ├── CODEOWNERS ├── FULL_README.md ├── LICENSE ├── MANIFEST.in ├── README.md ├── SECURITY.md ├── download.sh ├── examples_scaling ├── downstream │ ├── DOWNSTREAM.md │ ├── downstream_arm.sh │ ├── downstream_mdm.sh │ ├── downstream_mdm_mc.sh │ └── downstream_mdm_onlygen.sh ├── evaluation │ ├── EVAL.md │ ├── arm │ │ └── example_evaluation.sh │ └── mdm │ │ └── example_evaluation.sh ├── quickstart │ ├── eval_ar.sh │ ├── eval_mdm.sh │ ├── train_ar.sh │ └── train_mdm.sh └── training │ ├── TRAIN.md │ ├── arm │ ├── example_multi_node_training.sh │ ├── example_training.sh │ ├── example_training_chinchilla.sh │ └── example_training_data_value.sh │ ├── launch.sh │ └── mdm │ ├── example_multi_node_training.sh │ ├── example_training.sh │ ├── example_training_chinchilla.sh │ └── example_training_data_value.sh ├── install.sh ├── megatron ├── __init__.py ├── arguments.py ├── checkpointing.py ├── core │ ├── README.md │ ├── __init__.py │ ├── enums.py │ ├── fusions │ │ ├── __init__.py │ │ ├── fused_bias_dropout.py │ │ ├── fused_bias_gelu.py │ │ ├── fused_layer_norm.py │ │ └── fused_softmax.py │ ├── model_parallel_config.py │ ├── models │ │ ├── __init__.py │ │ └── gpt │ │ │ ├── __init__.py │ │ │ ├── gpt_embedding.py │ │ │ └── gpt_model.py │ ├── package_info.py │ ├── parallel_state.py │ ├── pipeline_parallel │ │ ├── __init__.py │ │ ├── deepspeed_zbh1_engine.py │ │ ├── deepspeed_zbh1_schedule.py │ │ ├── p2p_communication.py │ │ └── schedules.py │ ├── requirements.txt │ ├── sequence_parallel │ │ ├── __init__.py │ │ └── cross_entropy.py │ ├── tensor_parallel │ │ ├── __init__.py │ │ ├── cross_entropy.py │ │ ├── data.py │ │ ├── layers.py │ │ ├── mappings.py │ │ ├── random.py │ │ ├── utils.py │ │ └── weight_grad_store.py │ ├── transformer │ │ ├── __init__.py │ │ ├── attention.py │ │ ├── core_attention.py │ │ ├── custom_layers │ │ │ └── transformer_engine.py │ │ ├── enums.py │ │ ├── mlp.py │ │ ├── module.py │ │ ├── transformer_block.py │ │ ├── transformer_config.py │ │ ├── transformer_layer.py │ │ └── utils.py │ └── utils.py ├── data │ ├── Makefile │ ├── __init__.py │ ├── autoaugment.py │ ├── bert_dataset.py │ ├── biencoder_dataset_utils.py │ ├── blendable_dataset.py │ ├── data_samplers.py │ ├── dataset_utils.py │ ├── emotion_dataset.py │ ├── gpt_dataset.py │ ├── helpers.cpp │ ├── ict_dataset.py │ ├── image_folder.py │ ├── indexed_dataset.py │ ├── orqa_wiki_dataset.py │ ├── prompt_dataset.py │ ├── realm_dataset_utils.py │ ├── realm_index.py │ ├── t5_dataset.py │ ├── test │ │ ├── test_indexed_dataset.py │ │ └── test_preprocess_data.sh │ └── vit_dataset.py ├── dist_signal_handler.py ├── enums.py ├── fp16_deprecated │ └── loss_scaler.py ├── fused_kernels │ ├── __init__.py │ ├── compat.h │ ├── scaled_masked_softmax.cpp │ ├── scaled_masked_softmax.h │ ├── scaled_masked_softmax_cuda.cu │ ├── scaled_softmax.cpp │ ├── scaled_softmax_cuda.cu │ ├── scaled_upper_triang_masked_softmax.cpp │ ├── scaled_upper_triang_masked_softmax.h │ ├── scaled_upper_triang_masked_softmax_cuda.cu │ ├── tests │ │ ├── __init__.py │ │ └── test_fused_kernels.py │ └── type_shim.h ├── global_vars.py ├── indexer.py ├── initialize.py ├── memory.py ├── microbatches.py ├── model │ ├── __init__.py │ ├── bert_model.py │ ├── biencoder_model.py │ ├── classification.py │ ├── diff_gpt_model.py │ ├── distributed.py │ ├── enums.py │ ├── fused_bias_gelu.py │ ├── fused_layer_norm.py │ ├── fused_rmsnorm.py │ ├── fused_softmax.py │ ├── gpt_model.py │ ├── language_model.py │ ├── layernorm.py │ ├── module.py │ ├── multiple_choice.py │ ├── realm_model.py │ ├── rmsnorm.py │ ├── rotary_pos_embedding.py │ ├── t5_model.py │ ├── target_gpt_model.py │ ├── transformer.py │ ├── utils.py │ └── vision │ │ ├── __init__.py │ │ ├── classification.py │ │ ├── dino.py │ │ ├── esvit_swin_backbone.py │ │ ├── inpainting.py │ │ ├── knn_monitor.py │ │ ├── mit_backbone.py │ │ ├── swin_backbone.py │ │ ├── utils.py │ │ └── vit_backbone.py ├── mpu │ └── tests │ │ ├── __init__.py │ │ ├── commons.py │ │ ├── test_cross_entropy.py │ │ ├── test_data.py │ │ ├── test_initialize.py │ │ ├── test_layers.py │ │ └── test_random.py ├── optimizer │ ├── __init__.py │ ├── clip_grads.py │ ├── distrib_optimizer.py │ ├── grad_scaler.py │ └── optimizer.py ├── optimizer_param_scheduler.py ├── p2p_communication.py ├── profiler.py ├── static │ └── index.html ├── text_generation │ ├── __init__.py │ ├── api.py │ ├── beam_utils.py │ ├── communication.py │ ├── forward_step.py │ ├── generation.py │ ├── sampling.py │ └── tokenization.py ├── text_generation_server.py ├── text_generation_utils.py ├── timers.py ├── tokenizer │ ├── __init__.py │ ├── bert_tokenization.py │ ├── gpt2_tokenization.py │ └── tokenizer.py ├── training.py └── utils.py ├── pretrain_diff_gpt.py ├── pretrain_gpt.py ├── preview.png ├── requirements.txt ├── setup.py ├── tasks ├── data_utils.py ├── ensemble_classifier.py ├── eval_harness │ ├── download.py │ ├── evaluate.py │ ├── evaluate_diff.py │ └── report-to-csv.py ├── eval_utils.py ├── finetune_utils.py ├── glue │ ├── cola.py │ ├── data.py │ ├── finetune.py │ ├── mnli.py │ ├── mrpc.py │ ├── qnli.py │ ├── qqp.py │ ├── rte.py │ ├── sst2.py │ └── stsb.py ├── main.py ├── msdp │ ├── README.md │ ├── evaluate.py │ ├── main.py │ ├── metrics.py │ ├── preprocessing.py │ └── prompt.py ├── orqa │ ├── README.md │ ├── evaluate_orqa.py │ ├── evaluate_utils.py │ ├── supervised │ │ ├── data.py │ │ ├── eval_utils.py │ │ └── finetune.py │ └── unsupervised │ │ ├── nq.py │ │ ├── qa_utils.py │ │ └── tokenizers.py ├── race │ ├── data.py │ └── finetune.py ├── vision │ ├── classification │ │ ├── classification.py │ │ └── eval_utils.py │ ├── finetune_utils.py │ ├── main.py │ └── segmentation │ │ ├── cityscapes.py │ │ ├── data.py │ │ ├── finetune_segformer.py │ │ ├── finetune_setr.py │ │ ├── metrics.py │ │ ├── seg_heads.py │ │ ├── seg_models.py │ │ ├── transforms.py │ │ └── utils.py └── zeroshot_gpt │ ├── datasets.py │ ├── detokenizer.py │ └── evaluate.py ├── tests ├── __init__.py ├── conftest.py ├── functional_tests │ ├── __init__.py │ ├── python_test_utils │ │ ├── __init__.py │ │ ├── check_slurm_job_completion.py │ │ ├── get_test_results_from_tensorboard_logs.py │ │ ├── test_ci_pipeline.py │ │ └── test_resume_checkpoint_pipeline.py │ ├── shell_test_utils │ │ └── jobwait.sh │ ├── test_results │ │ ├── bert │ │ │ ├── bert_tp1_pp2_1nodes_50steps.json │ │ │ ├── bert_tp1_pp4_1nodes_50steps.json │ │ │ ├── bert_tp2_pp2_1nodes_50steps.json │ │ │ └── bert_tp4_pp1_1nodes_50steps.json │ │ └── gpt3 │ │ │ ├── gpt3_tp1_pp2_1nodes_50steps.json │ │ │ ├── gpt3_tp1_pp4_1nodes_50steps.json │ │ │ ├── gpt3_tp2_pp2_1nodes_50steps.json │ │ │ └── gpt3_tp4_pp1_1nodes_50steps.json │ └── test_scripts │ │ ├── bert │ │ ├── pretrain_bert_distributed_resume_checkpoint_test.sh │ │ ├── pretrain_bert_distributed_test.sh │ │ ├── sbatch_bert_distributed_resume_checkpoint_test.sh │ │ └── sbatch_bert_distributed_test.sh │ │ └── gpt3 │ │ ├── pretrain_gpt3_distributed_resume_checkpoint_test.sh │ │ ├── pretrain_gpt3_distributed_test.sh │ │ ├── sbatch_gpt3_distributed_resume_checkpoint_test.sh │ │ └── sbatch_gpt3_distributed_test.sh ├── models │ ├── __init__.py │ ├── test_gpt_embedding.py │ └── test_gpt_model.py ├── pipeline_parallel │ ├── __init__.py │ └── test_schedules.py ├── run_megatron.py ├── tensor_parallel │ └── __int__.py ├── test_megatron.py ├── transformer │ ├── __init__.py │ ├── test_core_attention.py │ ├── test_module.py │ ├── test_parallel_attention.py │ ├── test_parallel_mlp.py │ ├── test_parallel_transformer_block.py │ ├── test_parallel_transformer_layer.py │ └── test_transformer_config.py └── unit_tests │ ├── __init__.py │ ├── tensor_parallel │ ├── test_cross_entropy.py │ ├── test_data.py │ ├── test_mappings.py │ ├── test_random.py │ └── test_tensor_parallel_utils.py │ ├── test_basic.py │ ├── test_parallel_state.py │ ├── test_utilities.py │ └── test_utils.py ├── tools ├── __init__.py ├── bert_embedding │ ├── __init__.py │ ├── dataset.py │ ├── embed.py │ ├── external_libs.py │ ├── huggingface.py │ └── utils.py ├── checkpoint_loader_megatron.py ├── checkpoint_saver_megatron.py ├── checkpoint_util.py ├── convert_checkpoint │ ├── README.md │ ├── deepspeed_checkpoint.py │ ├── deepspeed_to_megatron.py │ ├── deepspeed_to_transformers.py │ ├── inspect_checkpoint.py │ └── inspect_deepspeed_checkpoint.py ├── generate_samples_gpt.py ├── hf2megads_weight_converter.py ├── linter.py ├── merge_datasets.py ├── openwebtext │ ├── README.md │ ├── add_id.py │ ├── blacklist_urls.py │ ├── cleanup_dataset.py │ ├── cleanup_fix_dataset.py │ ├── filter_ngrams.py │ ├── find_duplicates.py │ ├── group_duplicate_url.py │ ├── merge_jsons.py │ └── remove_group_duplicates.py ├── preprocess_data.py ├── preprocess_data_many_cores.py ├── preprocess_data_nmt.py ├── retro │ ├── README.md │ ├── __init__.py │ ├── cli │ │ ├── __init__.py │ │ ├── __main__.py │ │ └── cli.py │ ├── db │ │ ├── __init__.py │ │ ├── build.py │ │ ├── dataset.py │ │ └── utils.py │ ├── examples │ │ ├── get_dataset_configs.sh │ │ ├── get_preprocess_cmd.sh │ │ ├── preprocess_data.sh │ │ └── pretrain_model.sh │ ├── external_libs.py │ ├── index │ │ ├── __init__.py │ │ ├── build.py │ │ ├── factory.py │ │ ├── index.py │ │ ├── indexes │ │ │ ├── __init__.py │ │ │ ├── faiss_base.py │ │ │ └── faiss_par_add.py │ │ └── utils.py │ ├── main.py │ ├── query │ │ ├── __init__.py │ │ ├── chunk_dataset.py │ │ ├── query.py │ │ ├── retro_dataset.py │ │ └── utils.py │ └── utils.py ├── run_text_generation_server.py └── text_generation_cli.py └── utils ├── data ├── gpt2-merges.txt └── gpt2-vocab.json ├── datapaths ├── example_dataset.txt ├── train_c4_100m.txt └── val_c4.txt ├── epoch_tokens.sh └── model_params.sh /.coveragerc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/.coveragerc -------------------------------------------------------------------------------- /.github/workflows/python.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/.github/workflows/python.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/.gitignore -------------------------------------------------------------------------------- /.gitlab-ci.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/.gitlab-ci.yml -------------------------------------------------------------------------------- /CODEOWNERS: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/CODEOWNERS -------------------------------------------------------------------------------- /FULL_README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/FULL_README.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/LICENSE -------------------------------------------------------------------------------- /MANIFEST.in: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/MANIFEST.in -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/README.md -------------------------------------------------------------------------------- /SECURITY.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/SECURITY.md -------------------------------------------------------------------------------- /download.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/download.sh -------------------------------------------------------------------------------- /examples_scaling/downstream/DOWNSTREAM.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/downstream/DOWNSTREAM.md -------------------------------------------------------------------------------- /examples_scaling/downstream/downstream_arm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/downstream/downstream_arm.sh -------------------------------------------------------------------------------- /examples_scaling/downstream/downstream_mdm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/downstream/downstream_mdm.sh -------------------------------------------------------------------------------- /examples_scaling/downstream/downstream_mdm_mc.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/downstream/downstream_mdm_mc.sh -------------------------------------------------------------------------------- /examples_scaling/downstream/downstream_mdm_onlygen.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/downstream/downstream_mdm_onlygen.sh -------------------------------------------------------------------------------- /examples_scaling/evaluation/EVAL.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/evaluation/EVAL.md -------------------------------------------------------------------------------- /examples_scaling/evaluation/arm/example_evaluation.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/evaluation/arm/example_evaluation.sh -------------------------------------------------------------------------------- /examples_scaling/evaluation/mdm/example_evaluation.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/evaluation/mdm/example_evaluation.sh -------------------------------------------------------------------------------- /examples_scaling/quickstart/eval_ar.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/quickstart/eval_ar.sh -------------------------------------------------------------------------------- /examples_scaling/quickstart/eval_mdm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/quickstart/eval_mdm.sh -------------------------------------------------------------------------------- /examples_scaling/quickstart/train_ar.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/quickstart/train_ar.sh -------------------------------------------------------------------------------- /examples_scaling/quickstart/train_mdm.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/quickstart/train_mdm.sh -------------------------------------------------------------------------------- /examples_scaling/training/TRAIN.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/TRAIN.md -------------------------------------------------------------------------------- /examples_scaling/training/arm/example_multi_node_training.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/arm/example_multi_node_training.sh -------------------------------------------------------------------------------- /examples_scaling/training/arm/example_training.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/arm/example_training.sh -------------------------------------------------------------------------------- /examples_scaling/training/arm/example_training_chinchilla.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/arm/example_training_chinchilla.sh -------------------------------------------------------------------------------- /examples_scaling/training/arm/example_training_data_value.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/arm/example_training_data_value.sh -------------------------------------------------------------------------------- /examples_scaling/training/launch.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/launch.sh -------------------------------------------------------------------------------- /examples_scaling/training/mdm/example_multi_node_training.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/mdm/example_multi_node_training.sh -------------------------------------------------------------------------------- /examples_scaling/training/mdm/example_training.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/mdm/example_training.sh -------------------------------------------------------------------------------- /examples_scaling/training/mdm/example_training_chinchilla.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/mdm/example_training_chinchilla.sh -------------------------------------------------------------------------------- /examples_scaling/training/mdm/example_training_data_value.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/examples_scaling/training/mdm/example_training_data_value.sh -------------------------------------------------------------------------------- /install.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/install.sh -------------------------------------------------------------------------------- /megatron/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/__init__.py -------------------------------------------------------------------------------- /megatron/arguments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/arguments.py -------------------------------------------------------------------------------- /megatron/checkpointing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/checkpointing.py -------------------------------------------------------------------------------- /megatron/core/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/README.md -------------------------------------------------------------------------------- /megatron/core/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/__init__.py -------------------------------------------------------------------------------- /megatron/core/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/enums.py -------------------------------------------------------------------------------- /megatron/core/fusions/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /megatron/core/fusions/fused_bias_dropout.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/fusions/fused_bias_dropout.py -------------------------------------------------------------------------------- /megatron/core/fusions/fused_bias_gelu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/fusions/fused_bias_gelu.py -------------------------------------------------------------------------------- /megatron/core/fusions/fused_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/fusions/fused_layer_norm.py -------------------------------------------------------------------------------- /megatron/core/fusions/fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/fusions/fused_softmax.py -------------------------------------------------------------------------------- /megatron/core/model_parallel_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/model_parallel_config.py -------------------------------------------------------------------------------- /megatron/core/models/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /megatron/core/models/gpt/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/models/gpt/__init__.py -------------------------------------------------------------------------------- /megatron/core/models/gpt/gpt_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/models/gpt/gpt_embedding.py -------------------------------------------------------------------------------- /megatron/core/models/gpt/gpt_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/models/gpt/gpt_model.py -------------------------------------------------------------------------------- /megatron/core/package_info.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/package_info.py -------------------------------------------------------------------------------- /megatron/core/parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/parallel_state.py -------------------------------------------------------------------------------- /megatron/core/pipeline_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/pipeline_parallel/__init__.py -------------------------------------------------------------------------------- /megatron/core/pipeline_parallel/deepspeed_zbh1_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/pipeline_parallel/deepspeed_zbh1_engine.py -------------------------------------------------------------------------------- /megatron/core/pipeline_parallel/deepspeed_zbh1_schedule.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/pipeline_parallel/deepspeed_zbh1_schedule.py -------------------------------------------------------------------------------- /megatron/core/pipeline_parallel/p2p_communication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/pipeline_parallel/p2p_communication.py -------------------------------------------------------------------------------- /megatron/core/pipeline_parallel/schedules.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/pipeline_parallel/schedules.py -------------------------------------------------------------------------------- /megatron/core/requirements.txt: -------------------------------------------------------------------------------- 1 | pybind11 2 | torch 3 | regex -------------------------------------------------------------------------------- /megatron/core/sequence_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/sequence_parallel/__init__.py -------------------------------------------------------------------------------- /megatron/core/sequence_parallel/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/sequence_parallel/cross_entropy.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/__init__.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/cross_entropy.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/data.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/layers.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/mappings.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/random.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/utils.py -------------------------------------------------------------------------------- /megatron/core/tensor_parallel/weight_grad_store.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/tensor_parallel/weight_grad_store.py -------------------------------------------------------------------------------- /megatron/core/transformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/__init__.py -------------------------------------------------------------------------------- /megatron/core/transformer/attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/attention.py -------------------------------------------------------------------------------- /megatron/core/transformer/core_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/core_attention.py -------------------------------------------------------------------------------- /megatron/core/transformer/custom_layers/transformer_engine.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/custom_layers/transformer_engine.py -------------------------------------------------------------------------------- /megatron/core/transformer/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/enums.py -------------------------------------------------------------------------------- /megatron/core/transformer/mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/mlp.py -------------------------------------------------------------------------------- /megatron/core/transformer/module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/module.py -------------------------------------------------------------------------------- /megatron/core/transformer/transformer_block.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/transformer_block.py -------------------------------------------------------------------------------- /megatron/core/transformer/transformer_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/transformer_config.py -------------------------------------------------------------------------------- /megatron/core/transformer/transformer_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/transformer_layer.py -------------------------------------------------------------------------------- /megatron/core/transformer/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/transformer/utils.py -------------------------------------------------------------------------------- /megatron/core/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/core/utils.py -------------------------------------------------------------------------------- /megatron/data/Makefile: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/Makefile -------------------------------------------------------------------------------- /megatron/data/__init__.py: -------------------------------------------------------------------------------- 1 | from . import indexed_dataset 2 | -------------------------------------------------------------------------------- /megatron/data/autoaugment.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/autoaugment.py -------------------------------------------------------------------------------- /megatron/data/bert_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/bert_dataset.py -------------------------------------------------------------------------------- /megatron/data/biencoder_dataset_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/biencoder_dataset_utils.py -------------------------------------------------------------------------------- /megatron/data/blendable_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/blendable_dataset.py -------------------------------------------------------------------------------- /megatron/data/data_samplers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/data_samplers.py -------------------------------------------------------------------------------- /megatron/data/dataset_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/dataset_utils.py -------------------------------------------------------------------------------- /megatron/data/emotion_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/emotion_dataset.py -------------------------------------------------------------------------------- /megatron/data/gpt_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/gpt_dataset.py -------------------------------------------------------------------------------- /megatron/data/helpers.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/helpers.cpp -------------------------------------------------------------------------------- /megatron/data/ict_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/ict_dataset.py -------------------------------------------------------------------------------- /megatron/data/image_folder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/image_folder.py -------------------------------------------------------------------------------- /megatron/data/indexed_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/indexed_dataset.py -------------------------------------------------------------------------------- /megatron/data/orqa_wiki_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/orqa_wiki_dataset.py -------------------------------------------------------------------------------- /megatron/data/prompt_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/prompt_dataset.py -------------------------------------------------------------------------------- /megatron/data/realm_dataset_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/realm_dataset_utils.py -------------------------------------------------------------------------------- /megatron/data/realm_index.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/realm_index.py -------------------------------------------------------------------------------- /megatron/data/t5_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/t5_dataset.py -------------------------------------------------------------------------------- /megatron/data/test/test_indexed_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/test/test_indexed_dataset.py -------------------------------------------------------------------------------- /megatron/data/test/test_preprocess_data.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/test/test_preprocess_data.sh -------------------------------------------------------------------------------- /megatron/data/vit_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/data/vit_dataset.py -------------------------------------------------------------------------------- /megatron/dist_signal_handler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/dist_signal_handler.py -------------------------------------------------------------------------------- /megatron/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/enums.py -------------------------------------------------------------------------------- /megatron/fp16_deprecated/loss_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fp16_deprecated/loss_scaler.py -------------------------------------------------------------------------------- /megatron/fused_kernels/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/__init__.py -------------------------------------------------------------------------------- /megatron/fused_kernels/compat.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/compat.h -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_masked_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_masked_softmax.cpp -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_masked_softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_masked_softmax.h -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_masked_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_masked_softmax_cuda.cu -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_softmax.cpp -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_softmax_cuda.cu -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_upper_triang_masked_softmax.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_upper_triang_masked_softmax.cpp -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_upper_triang_masked_softmax.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_upper_triang_masked_softmax.h -------------------------------------------------------------------------------- /megatron/fused_kernels/scaled_upper_triang_masked_softmax_cuda.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/scaled_upper_triang_masked_softmax_cuda.cu -------------------------------------------------------------------------------- /megatron/fused_kernels/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /megatron/fused_kernels/tests/test_fused_kernels.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/tests/test_fused_kernels.py -------------------------------------------------------------------------------- /megatron/fused_kernels/type_shim.h: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/fused_kernels/type_shim.h -------------------------------------------------------------------------------- /megatron/global_vars.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/global_vars.py -------------------------------------------------------------------------------- /megatron/indexer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/indexer.py -------------------------------------------------------------------------------- /megatron/initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/initialize.py -------------------------------------------------------------------------------- /megatron/memory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/memory.py -------------------------------------------------------------------------------- /megatron/microbatches.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/microbatches.py -------------------------------------------------------------------------------- /megatron/model/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/__init__.py -------------------------------------------------------------------------------- /megatron/model/bert_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/bert_model.py -------------------------------------------------------------------------------- /megatron/model/biencoder_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/biencoder_model.py -------------------------------------------------------------------------------- /megatron/model/classification.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/classification.py -------------------------------------------------------------------------------- /megatron/model/diff_gpt_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/diff_gpt_model.py -------------------------------------------------------------------------------- /megatron/model/distributed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/distributed.py -------------------------------------------------------------------------------- /megatron/model/enums.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/enums.py -------------------------------------------------------------------------------- /megatron/model/fused_bias_gelu.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/fused_bias_gelu.py -------------------------------------------------------------------------------- /megatron/model/fused_layer_norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/fused_layer_norm.py -------------------------------------------------------------------------------- /megatron/model/fused_rmsnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/fused_rmsnorm.py -------------------------------------------------------------------------------- /megatron/model/fused_softmax.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/fused_softmax.py -------------------------------------------------------------------------------- /megatron/model/gpt_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/gpt_model.py -------------------------------------------------------------------------------- /megatron/model/language_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/language_model.py -------------------------------------------------------------------------------- /megatron/model/layernorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/layernorm.py -------------------------------------------------------------------------------- /megatron/model/module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/module.py -------------------------------------------------------------------------------- /megatron/model/multiple_choice.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/multiple_choice.py -------------------------------------------------------------------------------- /megatron/model/realm_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/realm_model.py -------------------------------------------------------------------------------- /megatron/model/rmsnorm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/rmsnorm.py -------------------------------------------------------------------------------- /megatron/model/rotary_pos_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/rotary_pos_embedding.py -------------------------------------------------------------------------------- /megatron/model/t5_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/t5_model.py -------------------------------------------------------------------------------- /megatron/model/target_gpt_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/target_gpt_model.py -------------------------------------------------------------------------------- /megatron/model/transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/transformer.py -------------------------------------------------------------------------------- /megatron/model/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/utils.py -------------------------------------------------------------------------------- /megatron/model/vision/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /megatron/model/vision/classification.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/classification.py -------------------------------------------------------------------------------- /megatron/model/vision/dino.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/dino.py -------------------------------------------------------------------------------- /megatron/model/vision/esvit_swin_backbone.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/esvit_swin_backbone.py -------------------------------------------------------------------------------- /megatron/model/vision/inpainting.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/inpainting.py -------------------------------------------------------------------------------- /megatron/model/vision/knn_monitor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/knn_monitor.py -------------------------------------------------------------------------------- /megatron/model/vision/mit_backbone.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/mit_backbone.py -------------------------------------------------------------------------------- /megatron/model/vision/swin_backbone.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/swin_backbone.py -------------------------------------------------------------------------------- /megatron/model/vision/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/utils.py -------------------------------------------------------------------------------- /megatron/model/vision/vit_backbone.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/model/vision/vit_backbone.py -------------------------------------------------------------------------------- /megatron/mpu/tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /megatron/mpu/tests/commons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/commons.py -------------------------------------------------------------------------------- /megatron/mpu/tests/test_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/test_cross_entropy.py -------------------------------------------------------------------------------- /megatron/mpu/tests/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/test_data.py -------------------------------------------------------------------------------- /megatron/mpu/tests/test_initialize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/test_initialize.py -------------------------------------------------------------------------------- /megatron/mpu/tests/test_layers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/test_layers.py -------------------------------------------------------------------------------- /megatron/mpu/tests/test_random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/mpu/tests/test_random.py -------------------------------------------------------------------------------- /megatron/optimizer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer/__init__.py -------------------------------------------------------------------------------- /megatron/optimizer/clip_grads.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer/clip_grads.py -------------------------------------------------------------------------------- /megatron/optimizer/distrib_optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer/distrib_optimizer.py -------------------------------------------------------------------------------- /megatron/optimizer/grad_scaler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer/grad_scaler.py -------------------------------------------------------------------------------- /megatron/optimizer/optimizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer/optimizer.py -------------------------------------------------------------------------------- /megatron/optimizer_param_scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/optimizer_param_scheduler.py -------------------------------------------------------------------------------- /megatron/p2p_communication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/p2p_communication.py -------------------------------------------------------------------------------- /megatron/profiler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/profiler.py -------------------------------------------------------------------------------- /megatron/static/index.html: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/static/index.html -------------------------------------------------------------------------------- /megatron/text_generation/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/__init__.py -------------------------------------------------------------------------------- /megatron/text_generation/api.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/api.py -------------------------------------------------------------------------------- /megatron/text_generation/beam_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/beam_utils.py -------------------------------------------------------------------------------- /megatron/text_generation/communication.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/communication.py -------------------------------------------------------------------------------- /megatron/text_generation/forward_step.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/forward_step.py -------------------------------------------------------------------------------- /megatron/text_generation/generation.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/generation.py -------------------------------------------------------------------------------- /megatron/text_generation/sampling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/sampling.py -------------------------------------------------------------------------------- /megatron/text_generation/tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation/tokenization.py -------------------------------------------------------------------------------- /megatron/text_generation_server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation_server.py -------------------------------------------------------------------------------- /megatron/text_generation_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/text_generation_utils.py -------------------------------------------------------------------------------- /megatron/timers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/timers.py -------------------------------------------------------------------------------- /megatron/tokenizer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/tokenizer/__init__.py -------------------------------------------------------------------------------- /megatron/tokenizer/bert_tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/tokenizer/bert_tokenization.py -------------------------------------------------------------------------------- /megatron/tokenizer/gpt2_tokenization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/tokenizer/gpt2_tokenization.py -------------------------------------------------------------------------------- /megatron/tokenizer/tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/tokenizer/tokenizer.py -------------------------------------------------------------------------------- /megatron/training.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/training.py -------------------------------------------------------------------------------- /megatron/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/megatron/utils.py -------------------------------------------------------------------------------- /pretrain_diff_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/pretrain_diff_gpt.py -------------------------------------------------------------------------------- /pretrain_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/pretrain_gpt.py -------------------------------------------------------------------------------- /preview.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/preview.png -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/requirements.txt -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/setup.py -------------------------------------------------------------------------------- /tasks/data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/data_utils.py -------------------------------------------------------------------------------- /tasks/ensemble_classifier.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/ensemble_classifier.py -------------------------------------------------------------------------------- /tasks/eval_harness/download.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/eval_harness/download.py -------------------------------------------------------------------------------- /tasks/eval_harness/evaluate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/eval_harness/evaluate.py -------------------------------------------------------------------------------- /tasks/eval_harness/evaluate_diff.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/eval_harness/evaluate_diff.py -------------------------------------------------------------------------------- /tasks/eval_harness/report-to-csv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/eval_harness/report-to-csv.py -------------------------------------------------------------------------------- /tasks/eval_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/eval_utils.py -------------------------------------------------------------------------------- /tasks/finetune_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/finetune_utils.py -------------------------------------------------------------------------------- /tasks/glue/cola.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/cola.py -------------------------------------------------------------------------------- /tasks/glue/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/data.py -------------------------------------------------------------------------------- /tasks/glue/finetune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/finetune.py -------------------------------------------------------------------------------- /tasks/glue/mnli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/mnli.py -------------------------------------------------------------------------------- /tasks/glue/mrpc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/mrpc.py -------------------------------------------------------------------------------- /tasks/glue/qnli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/qnli.py -------------------------------------------------------------------------------- /tasks/glue/qqp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/qqp.py -------------------------------------------------------------------------------- /tasks/glue/rte.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/rte.py -------------------------------------------------------------------------------- /tasks/glue/sst2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/sst2.py -------------------------------------------------------------------------------- /tasks/glue/stsb.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/glue/stsb.py -------------------------------------------------------------------------------- /tasks/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/main.py -------------------------------------------------------------------------------- /tasks/msdp/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/README.md -------------------------------------------------------------------------------- /tasks/msdp/evaluate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/evaluate.py -------------------------------------------------------------------------------- /tasks/msdp/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/main.py -------------------------------------------------------------------------------- /tasks/msdp/metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/metrics.py -------------------------------------------------------------------------------- /tasks/msdp/preprocessing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/preprocessing.py -------------------------------------------------------------------------------- /tasks/msdp/prompt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/msdp/prompt.py -------------------------------------------------------------------------------- /tasks/orqa/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/README.md -------------------------------------------------------------------------------- /tasks/orqa/evaluate_orqa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/evaluate_orqa.py -------------------------------------------------------------------------------- /tasks/orqa/evaluate_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/evaluate_utils.py -------------------------------------------------------------------------------- /tasks/orqa/supervised/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/supervised/data.py -------------------------------------------------------------------------------- /tasks/orqa/supervised/eval_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/supervised/eval_utils.py -------------------------------------------------------------------------------- /tasks/orqa/supervised/finetune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/supervised/finetune.py -------------------------------------------------------------------------------- /tasks/orqa/unsupervised/nq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/unsupervised/nq.py -------------------------------------------------------------------------------- /tasks/orqa/unsupervised/qa_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/unsupervised/qa_utils.py -------------------------------------------------------------------------------- /tasks/orqa/unsupervised/tokenizers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/orqa/unsupervised/tokenizers.py -------------------------------------------------------------------------------- /tasks/race/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/race/data.py -------------------------------------------------------------------------------- /tasks/race/finetune.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/race/finetune.py -------------------------------------------------------------------------------- /tasks/vision/classification/classification.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/classification/classification.py -------------------------------------------------------------------------------- /tasks/vision/classification/eval_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/classification/eval_utils.py -------------------------------------------------------------------------------- /tasks/vision/finetune_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/finetune_utils.py -------------------------------------------------------------------------------- /tasks/vision/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/main.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/cityscapes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/cityscapes.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/data.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/finetune_segformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/finetune_segformer.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/finetune_setr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/finetune_setr.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/metrics.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/seg_heads.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/seg_heads.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/seg_models.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/seg_models.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/transforms.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/transforms.py -------------------------------------------------------------------------------- /tasks/vision/segmentation/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/vision/segmentation/utils.py -------------------------------------------------------------------------------- /tasks/zeroshot_gpt/datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/zeroshot_gpt/datasets.py -------------------------------------------------------------------------------- /tasks/zeroshot_gpt/detokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/zeroshot_gpt/detokenizer.py -------------------------------------------------------------------------------- /tasks/zeroshot_gpt/evaluate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tasks/zeroshot_gpt/evaluate.py -------------------------------------------------------------------------------- /tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/conftest.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/conftest.py -------------------------------------------------------------------------------- /tests/functional_tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/functional_tests/python_test_utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/functional_tests/python_test_utils/check_slurm_job_completion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/python_test_utils/check_slurm_job_completion.py -------------------------------------------------------------------------------- /tests/functional_tests/python_test_utils/get_test_results_from_tensorboard_logs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/python_test_utils/get_test_results_from_tensorboard_logs.py -------------------------------------------------------------------------------- /tests/functional_tests/python_test_utils/test_ci_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/python_test_utils/test_ci_pipeline.py -------------------------------------------------------------------------------- /tests/functional_tests/python_test_utils/test_resume_checkpoint_pipeline.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/python_test_utils/test_resume_checkpoint_pipeline.py -------------------------------------------------------------------------------- /tests/functional_tests/shell_test_utils/jobwait.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/shell_test_utils/jobwait.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_results/bert/bert_tp1_pp2_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/bert/bert_tp1_pp2_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/bert/bert_tp1_pp4_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/bert/bert_tp1_pp4_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/bert/bert_tp2_pp2_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/bert/bert_tp2_pp2_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/bert/bert_tp4_pp1_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/bert/bert_tp4_pp1_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/gpt3/gpt3_tp1_pp2_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/gpt3/gpt3_tp1_pp2_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/gpt3/gpt3_tp1_pp4_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/gpt3/gpt3_tp1_pp4_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/gpt3/gpt3_tp2_pp2_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/gpt3/gpt3_tp2_pp2_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_results/gpt3/gpt3_tp4_pp1_1nodes_50steps.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_results/gpt3/gpt3_tp4_pp1_1nodes_50steps.json -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/bert/pretrain_bert_distributed_resume_checkpoint_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/bert/pretrain_bert_distributed_resume_checkpoint_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/bert/pretrain_bert_distributed_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/bert/pretrain_bert_distributed_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/bert/sbatch_bert_distributed_resume_checkpoint_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/bert/sbatch_bert_distributed_resume_checkpoint_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/bert/sbatch_bert_distributed_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/bert/sbatch_bert_distributed_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/gpt3/pretrain_gpt3_distributed_resume_checkpoint_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/gpt3/pretrain_gpt3_distributed_resume_checkpoint_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/gpt3/pretrain_gpt3_distributed_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/gpt3/pretrain_gpt3_distributed_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/gpt3/sbatch_gpt3_distributed_resume_checkpoint_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/gpt3/sbatch_gpt3_distributed_resume_checkpoint_test.sh -------------------------------------------------------------------------------- /tests/functional_tests/test_scripts/gpt3/sbatch_gpt3_distributed_test.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/functional_tests/test_scripts/gpt3/sbatch_gpt3_distributed_test.sh -------------------------------------------------------------------------------- /tests/models/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/models/test_gpt_embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/models/test_gpt_embedding.py -------------------------------------------------------------------------------- /tests/models/test_gpt_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/models/test_gpt_model.py -------------------------------------------------------------------------------- /tests/pipeline_parallel/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/pipeline_parallel/test_schedules.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/pipeline_parallel/test_schedules.py -------------------------------------------------------------------------------- /tests/run_megatron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/run_megatron.py -------------------------------------------------------------------------------- /tests/tensor_parallel/__int__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/test_megatron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/test_megatron.py -------------------------------------------------------------------------------- /tests/transformer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/transformer/test_core_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_core_attention.py -------------------------------------------------------------------------------- /tests/transformer/test_module.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_module.py -------------------------------------------------------------------------------- /tests/transformer/test_parallel_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_parallel_attention.py -------------------------------------------------------------------------------- /tests/transformer/test_parallel_mlp.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_parallel_mlp.py -------------------------------------------------------------------------------- /tests/transformer/test_parallel_transformer_block.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_parallel_transformer_block.py -------------------------------------------------------------------------------- /tests/transformer/test_parallel_transformer_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_parallel_transformer_layer.py -------------------------------------------------------------------------------- /tests/transformer/test_transformer_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/transformer/test_transformer_config.py -------------------------------------------------------------------------------- /tests/unit_tests/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tests/unit_tests/tensor_parallel/test_cross_entropy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/tensor_parallel/test_cross_entropy.py -------------------------------------------------------------------------------- /tests/unit_tests/tensor_parallel/test_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/tensor_parallel/test_data.py -------------------------------------------------------------------------------- /tests/unit_tests/tensor_parallel/test_mappings.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/tensor_parallel/test_mappings.py -------------------------------------------------------------------------------- /tests/unit_tests/tensor_parallel/test_random.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/tensor_parallel/test_random.py -------------------------------------------------------------------------------- /tests/unit_tests/tensor_parallel/test_tensor_parallel_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/tensor_parallel/test_tensor_parallel_utils.py -------------------------------------------------------------------------------- /tests/unit_tests/test_basic.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/test_basic.py -------------------------------------------------------------------------------- /tests/unit_tests/test_parallel_state.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/test_parallel_state.py -------------------------------------------------------------------------------- /tests/unit_tests/test_utilities.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/test_utilities.py -------------------------------------------------------------------------------- /tests/unit_tests/test_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tests/unit_tests/test_utils.py -------------------------------------------------------------------------------- /tools/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tools/bert_embedding/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/__init__.py -------------------------------------------------------------------------------- /tools/bert_embedding/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/dataset.py -------------------------------------------------------------------------------- /tools/bert_embedding/embed.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/embed.py -------------------------------------------------------------------------------- /tools/bert_embedding/external_libs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/external_libs.py -------------------------------------------------------------------------------- /tools/bert_embedding/huggingface.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/huggingface.py -------------------------------------------------------------------------------- /tools/bert_embedding/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/bert_embedding/utils.py -------------------------------------------------------------------------------- /tools/checkpoint_loader_megatron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/checkpoint_loader_megatron.py -------------------------------------------------------------------------------- /tools/checkpoint_saver_megatron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/checkpoint_saver_megatron.py -------------------------------------------------------------------------------- /tools/checkpoint_util.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/checkpoint_util.py -------------------------------------------------------------------------------- /tools/convert_checkpoint/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/README.md -------------------------------------------------------------------------------- /tools/convert_checkpoint/deepspeed_checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/deepspeed_checkpoint.py -------------------------------------------------------------------------------- /tools/convert_checkpoint/deepspeed_to_megatron.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/deepspeed_to_megatron.py -------------------------------------------------------------------------------- /tools/convert_checkpoint/deepspeed_to_transformers.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/deepspeed_to_transformers.py -------------------------------------------------------------------------------- /tools/convert_checkpoint/inspect_checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/inspect_checkpoint.py -------------------------------------------------------------------------------- /tools/convert_checkpoint/inspect_deepspeed_checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/convert_checkpoint/inspect_deepspeed_checkpoint.py -------------------------------------------------------------------------------- /tools/generate_samples_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/generate_samples_gpt.py -------------------------------------------------------------------------------- /tools/hf2megads_weight_converter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/hf2megads_weight_converter.py -------------------------------------------------------------------------------- /tools/linter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/linter.py -------------------------------------------------------------------------------- /tools/merge_datasets.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/merge_datasets.py -------------------------------------------------------------------------------- /tools/openwebtext/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/README.md -------------------------------------------------------------------------------- /tools/openwebtext/add_id.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/add_id.py -------------------------------------------------------------------------------- /tools/openwebtext/blacklist_urls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/blacklist_urls.py -------------------------------------------------------------------------------- /tools/openwebtext/cleanup_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/cleanup_dataset.py -------------------------------------------------------------------------------- /tools/openwebtext/cleanup_fix_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/cleanup_fix_dataset.py -------------------------------------------------------------------------------- /tools/openwebtext/filter_ngrams.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/filter_ngrams.py -------------------------------------------------------------------------------- /tools/openwebtext/find_duplicates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/find_duplicates.py -------------------------------------------------------------------------------- /tools/openwebtext/group_duplicate_url.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/group_duplicate_url.py -------------------------------------------------------------------------------- /tools/openwebtext/merge_jsons.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/merge_jsons.py -------------------------------------------------------------------------------- /tools/openwebtext/remove_group_duplicates.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/openwebtext/remove_group_duplicates.py -------------------------------------------------------------------------------- /tools/preprocess_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/preprocess_data.py -------------------------------------------------------------------------------- /tools/preprocess_data_many_cores.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/preprocess_data_many_cores.py -------------------------------------------------------------------------------- /tools/preprocess_data_nmt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/preprocess_data_nmt.py -------------------------------------------------------------------------------- /tools/retro/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/README.md -------------------------------------------------------------------------------- /tools/retro/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /tools/retro/cli/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/cli/__init__.py -------------------------------------------------------------------------------- /tools/retro/cli/__main__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/cli/__main__.py -------------------------------------------------------------------------------- /tools/retro/cli/cli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/cli/cli.py -------------------------------------------------------------------------------- /tools/retro/db/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/db/__init__.py -------------------------------------------------------------------------------- /tools/retro/db/build.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/db/build.py -------------------------------------------------------------------------------- /tools/retro/db/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/db/dataset.py -------------------------------------------------------------------------------- /tools/retro/db/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/db/utils.py -------------------------------------------------------------------------------- /tools/retro/examples/get_dataset_configs.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/examples/get_dataset_configs.sh -------------------------------------------------------------------------------- /tools/retro/examples/get_preprocess_cmd.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/examples/get_preprocess_cmd.sh -------------------------------------------------------------------------------- /tools/retro/examples/preprocess_data.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/examples/preprocess_data.sh -------------------------------------------------------------------------------- /tools/retro/examples/pretrain_model.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/examples/pretrain_model.sh -------------------------------------------------------------------------------- /tools/retro/external_libs.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/external_libs.py -------------------------------------------------------------------------------- /tools/retro/index/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/__init__.py -------------------------------------------------------------------------------- /tools/retro/index/build.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/build.py -------------------------------------------------------------------------------- /tools/retro/index/factory.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/factory.py -------------------------------------------------------------------------------- /tools/retro/index/index.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/index.py -------------------------------------------------------------------------------- /tools/retro/index/indexes/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/indexes/__init__.py -------------------------------------------------------------------------------- /tools/retro/index/indexes/faiss_base.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/indexes/faiss_base.py -------------------------------------------------------------------------------- /tools/retro/index/indexes/faiss_par_add.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/indexes/faiss_par_add.py -------------------------------------------------------------------------------- /tools/retro/index/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/index/utils.py -------------------------------------------------------------------------------- /tools/retro/main.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/main.py -------------------------------------------------------------------------------- /tools/retro/query/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/query/__init__.py -------------------------------------------------------------------------------- /tools/retro/query/chunk_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/query/chunk_dataset.py -------------------------------------------------------------------------------- /tools/retro/query/query.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/query/query.py -------------------------------------------------------------------------------- /tools/retro/query/retro_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/query/retro_dataset.py -------------------------------------------------------------------------------- /tools/retro/query/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/query/utils.py -------------------------------------------------------------------------------- /tools/retro/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/retro/utils.py -------------------------------------------------------------------------------- /tools/run_text_generation_server.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/run_text_generation_server.py -------------------------------------------------------------------------------- /tools/text_generation_cli.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/tools/text_generation_cli.py -------------------------------------------------------------------------------- /utils/data/gpt2-merges.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/utils/data/gpt2-merges.txt -------------------------------------------------------------------------------- /utils/data/gpt2-vocab.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/utils/data/gpt2-vocab.json -------------------------------------------------------------------------------- /utils/datapaths/example_dataset.txt: -------------------------------------------------------------------------------- 1 | "1.0 /path/to/your/trainset/" 2 | -------------------------------------------------------------------------------- /utils/datapaths/train_c4_100m.txt: -------------------------------------------------------------------------------- 1 | "1.0 data/gpt2tok_c4_100_text_document" -------------------------------------------------------------------------------- /utils/datapaths/val_c4.txt: -------------------------------------------------------------------------------- 1 | "1.0 data/gpt2tok_c4_val_text_document" -------------------------------------------------------------------------------- /utils/epoch_tokens.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/utils/epoch_tokens.sh -------------------------------------------------------------------------------- /utils/model_params.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/wmn-231314/diffusion-data-constraint/HEAD/utils/model_params.sh --------------------------------------------------------------------------------