├── README.md ├── demo.py ├── finetune ├── README.md ├── deepspeed │ └── cpu_offload.json ├── mc │ ├── README.md │ ├── data │ │ └── medqa_usmle_hf │ │ │ ├── dev.json │ │ │ ├── test.json │ │ │ └── train.json │ ├── preprocess_medqa.py │ ├── run_experiments.py │ └── run_multiple_choice.py ├── seqcls │ ├── README.md │ ├── data │ │ ├── bioasq_hf │ │ │ ├── dev.json │ │ │ ├── test.json │ │ │ └── train.json │ │ └── pubmedqa_hf │ │ │ ├── dev.json │ │ │ ├── test.json │ │ │ └── train.json │ ├── preprocess_blurb_seqcls.py │ └── run_seqcls_gpt.py ├── setup │ └── requirements.txt ├── textgen │ ├── data │ │ └── meqsum │ │ │ ├── test.source │ │ │ ├── test.target │ │ │ ├── train.source │ │ │ ├── train.target │ │ │ ├── val.source │ │ │ └── val.target │ └── gpt2 │ │ ├── finetune_for_summarization.py │ │ ├── generate_demo.py │ │ ├── run_generation_batch.py │ │ ├── sum_data_collator.py │ │ └── sum_dataset.py └── utils │ ├── custom_modeling_gpt2.py │ ├── custom_modeling_gpt_neo.py │ └── hf_flash_gpt_2.py └── tokenize └── train_bpe.py /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/README.md -------------------------------------------------------------------------------- /demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/demo.py -------------------------------------------------------------------------------- /finetune/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/README.md -------------------------------------------------------------------------------- /finetune/deepspeed/cpu_offload.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/deepspeed/cpu_offload.json -------------------------------------------------------------------------------- /finetune/mc/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/README.md -------------------------------------------------------------------------------- /finetune/mc/data/medqa_usmle_hf/dev.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/data/medqa_usmle_hf/dev.json -------------------------------------------------------------------------------- /finetune/mc/data/medqa_usmle_hf/test.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/data/medqa_usmle_hf/test.json -------------------------------------------------------------------------------- /finetune/mc/data/medqa_usmle_hf/train.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/data/medqa_usmle_hf/train.json -------------------------------------------------------------------------------- /finetune/mc/preprocess_medqa.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/preprocess_medqa.py -------------------------------------------------------------------------------- /finetune/mc/run_experiments.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/run_experiments.py -------------------------------------------------------------------------------- /finetune/mc/run_multiple_choice.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/mc/run_multiple_choice.py -------------------------------------------------------------------------------- /finetune/seqcls/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/README.md -------------------------------------------------------------------------------- /finetune/seqcls/data/bioasq_hf/dev.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/bioasq_hf/dev.json -------------------------------------------------------------------------------- /finetune/seqcls/data/bioasq_hf/test.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/bioasq_hf/test.json -------------------------------------------------------------------------------- /finetune/seqcls/data/bioasq_hf/train.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/bioasq_hf/train.json -------------------------------------------------------------------------------- /finetune/seqcls/data/pubmedqa_hf/dev.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/pubmedqa_hf/dev.json -------------------------------------------------------------------------------- /finetune/seqcls/data/pubmedqa_hf/test.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/pubmedqa_hf/test.json -------------------------------------------------------------------------------- /finetune/seqcls/data/pubmedqa_hf/train.json: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/data/pubmedqa_hf/train.json -------------------------------------------------------------------------------- /finetune/seqcls/preprocess_blurb_seqcls.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/preprocess_blurb_seqcls.py -------------------------------------------------------------------------------- /finetune/seqcls/run_seqcls_gpt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/seqcls/run_seqcls_gpt.py -------------------------------------------------------------------------------- /finetune/setup/requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/setup/requirements.txt -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/test.source: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/test.source -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/test.target: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/test.target -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/train.source: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/train.source -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/train.target: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/train.target -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/val.source: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/val.source -------------------------------------------------------------------------------- /finetune/textgen/data/meqsum/val.target: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/data/meqsum/val.target -------------------------------------------------------------------------------- /finetune/textgen/gpt2/finetune_for_summarization.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/gpt2/finetune_for_summarization.py -------------------------------------------------------------------------------- /finetune/textgen/gpt2/generate_demo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/gpt2/generate_demo.py -------------------------------------------------------------------------------- /finetune/textgen/gpt2/run_generation_batch.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/gpt2/run_generation_batch.py -------------------------------------------------------------------------------- /finetune/textgen/gpt2/sum_data_collator.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/gpt2/sum_data_collator.py -------------------------------------------------------------------------------- /finetune/textgen/gpt2/sum_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/textgen/gpt2/sum_dataset.py -------------------------------------------------------------------------------- /finetune/utils/custom_modeling_gpt2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/utils/custom_modeling_gpt2.py -------------------------------------------------------------------------------- /finetune/utils/custom_modeling_gpt_neo.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/utils/custom_modeling_gpt_neo.py -------------------------------------------------------------------------------- /finetune/utils/hf_flash_gpt_2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/finetune/utils/hf_flash_gpt_2.py -------------------------------------------------------------------------------- /tokenize/train_bpe.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/stanford-crfm/BioMedLM/HEAD/tokenize/train_bpe.py --------------------------------------------------------------------------------