alon-albalak / flad Goto Github PK

Hi!
Thanks for the great code
I was able to run the explore only baseline fairly easily but facing some issues with the exploit only baseline. I'll go over the issues I encountered and how I tried fixing them step by step.

1. Running the code directly led to an error on this line. I think this is due to the fact that instead of args, the attribute is being extracted from training_args where it does not exist
2. I make the necessary change but then I get an error for the path being None on the same line so I decide to add a path myself as the default path is None and the bash file does not provide a path. To do so I add the arg in the bash file so my bash file now looks like this -

for SEED in ${SEEDS[@]}; do
    for AUX_DATASET in ${AUX_DATASETS[@]}; do
        for MODEL in ${MODELS[@]}; do
        
            OUTPUT_DIR="outputs/train_logs/exploit_only/$SEED/$MODEL/${AUX_DATASET}/${TARGET_DATASET}/1000"
            PRECOMPUTED_WEIGHT_GRAD_SAVE_DIR="outputs/precomputed_weight_grads/exploit_only/$SEED/$MODEL/${AUX_DATASET}/${TARGET_DATASET}/1000"
            mkdir -p $OUTPUT_DIR

            echo $(date)
            echo "Running $SEED $MODEL $AUX_DATASET $TARGET_DATASET exploit only"
            echo "Saving log to ${OUTPUT_DIR}"

            CUDA_VISIBLE_DEVICES=$GPU python src/multirun_train_mixed.py \
                --seed $SEED \
                --target_dataset $TARGET_DATASET \
                --aux_dataset $AUX_DATASET \
                --model $MODEL \
                --weight_initialization_samples 1000 \
                --precomputed_weight_grad_save_dir $PRECOMPUTED_WEIGHT_GRAD_SAVE_DIR \
                > $OUTPUT_DIR/log.log 2> $OUTPUT_DIR/err.log
        done
    done
done

3. After doing this I get this error - Weight save file outputs/precomputed_weight_grads/exploit_only/42/google/t5-xl-lm-adapt/T0Mixture/copa/1000/initial_similarities/1000_copa_T0Mixture_google-t5-xl-lm-adapt_42.json does not exist. I looked into the code and noticed that the error is caused here however the file is assumed to be there apriori. How does one create this file?

This also extends to other runs such as the UCB1 run. There is a command to explicitly create the gradients for all datasets separately however I thought it would be run implicitly based on the README. Do let me know if I need to run it separately.