/KEPT

Primary LanguageJupyter NotebookMIT LicenseMIT

KEPT

PWC

This repository contains the implementation of our KEPT model on the auto icd coding task presented in EMNLP. This branch only contain code to experiment MIMIC-III-50 and MIMIC-III-rare50 in the paper. For MIMIC-III-full experiments, see the rerank300 branch.

Thanks to Zheng Yuan for opensourcing MSMN project, our evaluation code and data preprocsing step is heavily based on their repo.

Dependencies

  • Python 3
  • NumPy
  • PyTorch (currently tested on version 1.9.0+cu111)
  • Transformers (currently tested on version 4.16.2)
  • tqdm==4.62.2
  • ujson==5.3.0

Full environment setting is lised here and can be installed through:

conda env create -f conda-environment.yaml
conda activate ctorch191

Download / preprocess data

One need to obtain licences to download MIMIC-III dataset. Once you obtain the MIMIC-III dataset, please follow caml-mimic to preprocess the dataset. You should obtain train_full.csv, test_full.csv, dev_full.csv, train_50.csv, test_50.csv, dev_50.csv after preprocessing. Please put them under sample_data/mimic3. Then you should use preprocess/generate_data_new.ipynb for generating json format dataset for train/dev/test. A new data will be saved in ./sample_data/mimic3.

Modify constant

Modify constant.py : change DATA_DIR to where your preprocessed data located.

To enable wandb, modify wandb.init(project="PROJECTNAME", entity="WANDBACCOUNT") in run_coder.py.

Generate MIMIC-III-rare50 data

Run command below and rare50 data will be created like mimic3-50l_xxx.json and xxx_50l.csv. The ./sample_data/mimic3 folder will look something like this, without xxx_preds folders (they are prediction result for 2 stage reranker, see the correpdong branch to create those folders).

python collectrare50data.py

Hyperparameter and config setting to reproduce

If needed, see wandb.

Train and Eval

UMLS knowledge enhaced longformer is avail here.

To Train MIMIC-III 50 (2 A100 GPU):

CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nproc_per_node 2 --master_port 57666 run_coder.py \
                --ddp_find_unused_parameters False \
                --disable_tqdm True \
                --version mimic3-50 --model_name_or_path whaleloops/keptlongformer \
                --do_train --do_eval --max_seq_length 8192 \
                --per_device_train_batch_size 1 --per_device_eval_batch_size 2 \
                --learning_rate 1.5e-5 --weight_decay 1e-3 --adam_epsilon 1e-7 --num_train_epochs 8 \
                --evaluation_strategy epoch --save_strategy epoch \
                --logging_first_step --global_attention_strides 1 \
                --output_dir ./saved_models/longformer-original-clinical-prompt2alpha

To Train MIMIC-III 50 (2 V100 GPU), with slightly less accuracy from less global attention in the prompt as global_attention_strides=3. More details could be found in Table A.5:

CUDA_VISIBLE_DEVICES=2,3 python -m torch.distributed.launch --nproc_per_node 2 --master_port 57666 run_coder.py \
                --ddp_find_unused_parameters False \
                --disable_tqdm True \
                --version mimic3-50 --model_name_or_path whaleloops/keptlongformer \
                --do_train --do_eval --max_seq_length 8192 \
                --per_device_train_batch_size 1 --per_device_eval_batch_size 2 \
                --learning_rate 1.5e-5 --weight_decay 1e-3 --adam_epsilon 1e-7 --num_train_epochs 8 \
                --evaluation_strategy epoch --save_strategy epoch \
                --logging_first_step --global_attention_strides 3 \
                --output_dir OUTPUT_DIR

A finetuned model (where global_attention_strides=1) is available here. To eval MIMIC-III 50, change DOWNLOAD_MODEL_NAME_OR_PATH to the downloaded path:

CUDA_VISIBLE_DEVICES=0 python run_coder.py \
                --ddp_find_unused_parameters False \
                --disable_tqdm True \
                --version mimic3-50 --model_name_or_path DOWNLOAD_MODEL_NAME_OR_PATH \
                --do_eval --do_predict --max_seq_length 8192 \
                --per_device_train_batch_size 1 --per_device_eval_batch_size 2 \
                --learning_rate 1.5e-5 --weight_decay 1e-3 --adam_epsilon 1e-7 --num_train_epochs 8 \
                --evaluation_strategy epoch --save_strategy epoch \
                --logging_first_step --global_attention_strides 1 \
                --output_dir OUTPUT_DIR

To train and eval MIMIC-III rare50 by tuning only bias term and lm_head like BitFit, change DOWNLOAD_MODEL_NAME_OR_PATH to the downloaded path:

CUDA_VISIBLE_DEVICES=2,3 python -m torch.distributed.launch --nproc_per_node 2 --master_port 57666 run_coder.py \
                --ddp_find_unused_parameters False \
                --finetune_terms "bias;lm_head" \
                --disable_tqdm True \
                --version mimic3-50l --model_name_or_path DOWNLOAD_MODEL_NAME_OR_PATH \
                --do_train --do_eval --max_seq_length 8192 \
                --per_device_train_batch_size 1 --per_device_eval_batch_size 2 \
                --learning_rate 1e-3 --num_train_epochs 8 \
                --evaluation_strategy epoch --save_strategy epoch \
                --load_best_model_at_end True --metric_for_best_model eval_f1_macro --greater_is_better True \
                --logging_first_step --global_attention_strides 1 \
                --output_dir OUTPUT_DIR

Citation

@inproceedings{yang-etal-2022-knowledge-injected,
    title = "Knowledge Injected Prompt Based Fine-tuning for Multi-label Few-shot {ICD} Coding",
    author = "Yang, Zhichao  and
      Wang, Shufan  and
      Rawat, Bhanu Pratap Singh  and
      Mitra, Avijit  and
      Yu, Hong",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
    month = dec,
    year = "2022",
    address = "Abu Dhabi, United Arab Emirates",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2022.findings-emnlp.127",
    pages = "1767--1781",
}

License

See the LICENSE file for more details.

Branches

Supported branches

  • main: KEPTLongformer on MIMIC-III-50 and MIMIC-III-rare50.
  • rerank300: KEPTLongformer as reranker on MIMIC-III-full.

Deprecated branches

  • cls50: 1 global attention per label code description
  • syn_explicit: Adding synonyms like MSMN per label code description