Unable to run the FEVER task classifier scripts (LIME-based masker)

[JudePark96]

Hi,

To reproduce an experiments from the scratch that you've done, I was trying to run the LIME-based masker w/ bash scripts following as:

bash scripts/finetune_fever.sh 5e-5 fever_sub genre_50_2 true

finetune_fever.sh

lr=$1
reader=$2
ns=$3
dpr=$4

python -m fec.modelling.finetune_fever \
  --model_name_or_path bert-base-uncased \
  --learning_rate $lr \
  --num_train_epochs 3 \
  --output_dir output/fever/reader=${reader},lr=${lr},data=${dpr},negative_sampling=${ns}/seed=1/  \
  --train_file resources/retrieval/genre_50/${dpr}_train.jsonl \
  --val_file resources/retrieval/genre_50/${dpr}_dev.jsonl \
  --do_train \
  --train_batch_size 16 \
  --eval_batch_size 16 \
  --reader $reader \
  --negative_sampling ${ns} \
  --seed 1

But "ModuleNotFoundError" (fec.modelling.finetune_fever) comes out, and I also have tried to find fec module, but I couldn't find.

If you can, I hope you to check the bash script is wrong or not.

Thank you for your paper and code!

[j6mes]

Hi, please check the python path as this needs to contain the src/ folder which contains the fec module (run export PYTHONPATH=src)

[JudePark96]

@j6mes thank you for your reply.

I followed your environment set up process (create new conda repo, export PYTHONPATH=src), but it didn't work. But I checked bash scripts/finetune_supervised.sh t5-base 1e-4 true false all (Fully-supervised ceiling) is working. So I think it is not my environment problem.

Can you please check your repository contains fec module or not?

Thanks.

[j6mes]

Ah! Looks like it was an error in the script (when i made it public, i renamed the module from fec to error_correction)

I've updated the script now.

[JudePark96]

I checked update! Thanks!

[JudePark96]

@j6mes I think finetune_fever module doesn't exist. Can you please check again ?