Skip to content

Latest commit

 

History

History
126 lines (82 loc) · 3.33 KB

README.md

File metadata and controls

126 lines (82 loc) · 3.33 KB

GenBioEL

Here are reported the steps necessary to run GenBioEL with BELB data.

For each command make sure that (a) you have the correct virtual environment activated and (b) you are in the right directory. In the example the virtual environment name is signaled in brackets and after that there is the folder (repository) name.

1. Setup GenBioEL

You first need to clone the GenBioEL repository:

(belb) home $ git clone https://github.com/Yuanhy1997/GenBioEL

Then follow the instructions to setup the environment including obtaining the original data provided by the authors.

2. Sanity check

Place all the scripts you find in this folder into the BioSyn one.

Make sure that the script train_ncbi.sh (training with data provided by the authors) runs without issues (at least until it starts the actual training).

3. Convert BELB data

Run:

(belb) belb-benchmark $ python -m benchmark.genbioel.genbioel --run input --in_dir data_directory --out_dir genbioel_directory --belb belb_directory

This will convert BELB data into the format required by BioSyn and store it into in_dir.

4. GenBioEL preprocessing

Run:

(genbioel) GenBioEL $ ./create_belb_runs.sh data_directory

This will place each KB in the correct corpus directory and pre-process the data as required by GenBioEL.

Note that the path passed to --dir must be the same one passed to --in_dir in step 3.

5. Get BART

The training script downloads the BART weights at every call, which may throw this error if called repeatedly:

Traceback (most recent call last):
  File "/vol/home-vol3/wbi/gardasam/.venv/genbioel/lib/python3.9/site-packages/huggingface_hub/utils/_errors.py", line 259, in hf_raise_for_status
    response.raise_for_status()
  File "/vol/home-vol3/wbi/gardasam/.venv/genbioel/lib/python3.9/site-packages/requests/models.py", line 1021, in raise_for_status
    raise HTTPError(http_error_msg, response=self)
requests.exceptions.HTTPError: 504 Server Error: Gateway Time-out for url: https://huggingface.co/api/models/facebook/bart-large

To avoid this run:

(genbioel) GenBioEL $ git install lfs
(genbioel) GenBioEL $ git clone https://huggingface.co/facebook/bart-large

to save locally the BART weights and tokenizer.

6. Train

Edit the file train_belb.sh:

DIRECTORY="/path/to/genbioel/preprocessed/data"
# If you skipped step 6: 'facebook/bart-large' 
PLM="path/to/local/facebook/bart/weights"

DIRECTORY should point to the preprocessed data (steps 3 and 4) and PLM to the BART large model/tokenizer (step 5)

Run:

(genbioel) GenBioEL $ ./train_belb.sh

to train a GenBioEL model on each of the BELB corpora. Modify the CORPORA variable to specify only a subset.

7. Predict

You need to change this line

from:

pickle.dump([results, results_score], f)

to:

pickle.dump([results, cui_results], f)

This will avoid storing all beam search probabilities and save instead the name converted to entities atomic labels.

Run:

(genbioel) GenBioEL $ ./predict_belb.sh

8. Gather results

Run:

(belb-venv) belb-benchmark $ python -m benchmark.genbioel.genbioel --run output --in_dir data_directory --out_dir genbioel_directory --belb belb_directory