mntlra / DEXTER

This repository contains the source code of DEXTER, a system to automatically extract Gene-Disease Associations from biomedical abstracts.
0 stars 0 forks source link

Issue with Code Execution: Parsing Error #1

Closed sumit-kumar-9297 closed 6 months ago

sumit-kumar-9297 commented 6 months ago

Hello,

I trust this message finds you well.

I encountered an issue while attempting to execute the provided code. Upon execution, I generated a log file named "dexter.log," which indicates that although there are a total of 6189 documents, they are not being parsed for extracting relations as expected.

Could you kindly guide me in resolving this issue? I have attached the aforementioned log file for your reference.

Thank you very much for your attention to this matter.

Best regards, Sumit Kumar dexter.log

mntlra commented 6 months ago

Dear Kumar,

Thank you for reaching out.

I tested the code and did not encounter the issue you are experiencing. In particular, I cloned the repository, set a new conda environment, and ran the instruction provided in the README file and had no problem. I made some small adjustments to the code and the requirements list so I suggest you pull the repository again, check if your environment is set correctly, and follow the updated instructions to run the DEXTER_data. Make sure you have correctly installed the "en-sci-sm" model from Scispacy (https://allenai.github.io/scispacy/).

For reference, I attach my log file (file: "dexter_output.log") and the configuration file of my conda environment (file: "environment.txt"). If you want to build an identical conda environmental starting from the configuration file:

  1. Change the extension of the file from ".yml" to ".txt" (I had to save it as .txt to upload it here)
  2. Create a conda environment following the instructions at: creating an environment from an environment yml file

Let me know if this works.

Kind Regards, Laura Menotti

dexter_output.log environment.txt

sumit-kumar-9297 commented 6 months ago

Dear Menotti,

Thanks for writing back with solution, this is now working better than before and documents are parsed successfully. I appreciate you time and consideration for this.

Regards, Sumit Kumar