XZhang97666

[BUG]

2

# export CUDA_VISIBLE_DEVICES=2,3 task=medqa_usmle_hf datadir=data/$task outdir=runs/$task/GPT2 mkdir -p $outdir seed=42 deepspeed --num_gpus 2 --num_nodes 1 run_multiple_choice.py --tokenizer_name stanford-crfm/pubmed_gpt_tokenizer --model_name_or_path "stanford-crfm/BioMedLM" \ --train_file ../../../SLMReason/data/MedQA/BertMC/train.json --validation_file ../../../SLMReason/data/MedQA/BertMC/validation.json \ --test_file ../../../SLMReason/data/MedQA/BertMC/test.json --do_train --do_eval...

Is there a no_trainer version for image pertaining

8

### Feature request I wonder if there is a run_no_trainer code for image pertaining? @NielsRogge ### Motivation I usually use no trainer in my code because the trainer cannot should...

How to get hint's raw text for DPR.

1

I found there is no raw text from DenseSearchResult. How could I get corresponding text from retrieved results.

Hi, I noticed that a lot of "Chat Doctor" appears in the outputs of HealthCareMagic-100k.json. For example: ""Hi thanks for contacting Chat Doctor ... Your brother have both hepatitis b...