Open zihui-debug opened 4 months ago
Thanks for the issue. I missed this in for former sharing and have uploaded into https://huggingface.co/datasets/chaoyi-wu/RadFM_data_csv/blob/main/paper_train.csv. Generally, we use all papers and just dismiss some papers used for generating the test set in other datasets to avoid potential data leakage.
Thanks for the issue. I missed this in for former sharing and have uploaded into https://huggingface.co/datasets/chaoyi-wu/RadFM_data_csv/blob/main/paper_train.csv. Generally, we use all papers and just dismiss some papers used for generating the test set in other datasets to avoid potential data leakage.
thanks a lot! In addition, could you share the csv data for multi-label task? I didn't find these files in the sharing zip file: chestxray_new.csv
pcxr_train_new.csv
mammo_train_new.csv
spinexr_train_new.csv
Hello, thanks for your contributory work! I find that there isn't a
paper_train.csv
in thedata_csv.zip
. Is the paper path in this csv file the same as the PMC-Inline text json file from your huggingface at https://huggingface.co/datasets/chaoyi-wu/PMC-Inline/tree/main ? If not, how can I get thepaper_train.csv
to perform pretraining?