search

srbhr / Resume-Matcher

Resume Matcher is an open source, free tool to improve your resume. It works by using language models to compare and rank resumes with job descriptions.
https://www.resumematcher.fyi/
Apache License 2.0
4.91k stars 2.1k forks source link

I am not able to see the experience and matching score in the json output or in the streamlit server #242

Open azarite123 opened 6 months ago

azarite123 commented 6 months ago

When I am running the code, it generates the JSON file for both JD and Resumes, But I am not able to see the %match of JD and Resume and I am not able to see the extraction experience from both JD and Resume, Could you help me with that issue.

Svetlana-T commented 6 months ago

What OS? What version of Resume Matcher? What versions of its dependencies? Did it work before, or is this the first install?

srbhr commented 6 months ago

Thanks, @Svetlana-T, for the questions. @azarite123, please let me know the OS you are using. Have you added the Qdrant and Cohere API keys?

azarite123 commented 6 months ago

Hi @Svetlana-T and @srbhr , My OS is windows 11 and i haven't added the Qdrant and cohere API keys, Could you guide me on where would i be getting the API keys and i have another question as well, Is it possible to modify or get the similarity score of resume and JD in a json than in a streamlit web app. If so, Please help me with that.

srbhr commented 6 months ago

i haven't added the Qdrant and cohere API keys, We have a new PR which is in testing @azarite123 once that's merged you will be able to use similarity score without the need of more APIs.

azarite123 commented 6 months ago

thankyou @srbhr , Could you tell me when possibly we would be seeing the new PR being merged?

srbhr commented 6 months ago

Hi @azarite123, This is the PR: https://github.com/srbhr/Resume-Matcher/pull/243 It's getting merged this Friday. There's some testing going on for this Branch. You may help in the testing for it. Please check PR comments for more information.

azarite123 commented 6 months ago

Yes , sure. I'll test and let you know. Thankyou for the prompt reply. If there are any queries ,i'll reach out to this thread

srbhr commented 6 months ago

@azarite123 please try it now. The score should be visible without any API keys.

azarite123 commented 6 months ago

Hey @srbhr, the similarity scoring part is working accurately but I am not able to see the experience keywords in the Json file like how many years of experience a candidate has for JD and Resume is not being displayed .

  1. Is this calculated based on the semantic similarity by Qdrant if not how is calculated?
  2. the machine learning model, is extracting all the field except the number of experiences image (7)
srbhr commented 6 months ago

It's Machine Learning code @azarite123

jnnla commented 1 month ago

Hello. I'm having this same problem it seems. Everything is getting parsed appropriately but there is no Vector Similarity Score in the streamlit app - only the fake resume bar-plots. I am on windows 11, have installed no additional api keys (and would need instruction on what steps to take and for what dependencies). I would love to see a similarity rating. I installed everything as instructed on the main page and then implemented the folder-fix to get my data processed correctly. Any idea on how to get the Vector Similarity Score working?