search

zhuemann / ConTEXTualSegmentation

7 stars 0 forks source link

Question about Code Reproduction Issue #1

Open AllenZzzzzzzz opened 1 year ago

AllenZzzzzzzz commented 1 year ago

Hi , I have a question about reproducing your code. I followed your README instructions to obtain the dataset successfully. However, I couldn't find two Excel files, "pneumothorax_with_multisegmentation_positive_text_df.xlsx" and "words_and_their_synonyms.xlsx," which seem to be essential for the code to run.

Could you kindly provide access to these files or include them in your repository? It would be immensely helpful for replicating your work accurately. If sharing the files is not possible, any guidance on how to configure the necessary data would be greatly appreciated.

Additionally, if there are any other important considerations or prerequisites to run your code successfully, please let me know.

Looking forward to your response.

Thanks.

zmwyl commented 1 month ago

Hello, could you please share this dataset? I look forward to your reply. Thank you!

zhuemann commented 1 month ago

Unfortunately it is not my dataset to share, you will have to sign a user agreement with the owner to get the download.

You can contact them here: https://figshare.com/articles/dataset/CANDID-PTX/14173982

zmwyl commented 1 month ago

Thank you for your reply!

I have obtained the dataset, but I am facing the same issue. I can't find the two Excel files, "pneumothorax_with_multisegmentation_positive_text_df.xlsx" and "words_and_their_synonyms.xlsx," which seem to be necessary for running the code.

I would like to know if there are any other important considerations or prerequisites for successfully running the code. I look forward to your reply. Thank you!

zhuemann commented 1 month ago

I have added the "words_and_their_synonyms.xlsx" but note that this did not improve scores in the paper. It actually hurt performance so this can be removed but I included it for reproducibility.

Additionally I cannot include the exact file "pneumothorax_with_multisegmentation_positive_text_df.xlsx" but have included the form in the file named "image_name_text_label_redacted.csv". You will need to join the csv with the one found after downloading the data to get the labels and text. I also seem to remember their RLE being a bit weird so double check they look correct.

zmwyl commented 1 month ago

Thank you very much for your reply! Thanks!