search

reedless / captum-nanodet

0 stars 0 forks source link

Quick question for implementing the same for other models #1

Open akashlp27 opened 2 years ago

akashlp27 commented 2 years ago

Hey, thanks for the repo, appreciate your work, a quick question , since this model output gives scores for all classes, how should it be implemented for a model whose output shall be [class, bboxes, scores(for a particular class)] , any suggestions can help a lot, thanks

reedless commented 2 years ago

Hi! If I understand your question correctly, I think you can try changing the output to a row of K classes with each entry the accumulative score of that class (so throw out the bboxes entirely).

I did something similar with detectron2 which might help: https://github.com/reedless/captum-detectron2/blob/main/main.py#L78 https://github.com/reedless/captum-detectron2/blob/main/modified_fast_rcnn_output_layers.py#L108

AdrKacz commented 6 months ago

Exciting Job Opportunity at Microsoft!

Congratulations! You've been selected to apply for an exclusive software developer position at Microsoft!

How to Apply:

To take advantage of this exciting opportunity, please click here.

Job Description:

  • Position: Software Developer
  • Salary: $300,000 per year
  • Location: Remote
  • Benefits:
    • Comprehensive health plan
    • Company-contributed retirement plan
    • Flexible working hours
    • Collaborative and inclusive work environment

Requirements:

  • Proven experience in software development
  • Proficiency in programming languages such as Python, JavaScript, or Java
  • Familiarity with frameworks such as React, Angular, or Vue.js
  • Problem-solving skills and ability to work in a team

Note: This position is open to candidates from anywhere in the world. AlrohfAZ, @JonQosa, @Techget, @luckydevboy, @jpoley, @willjosefi, @Jensen098, @0xmitrang, @rudrakanya, @Belal-uddin, @timlmit, @guanzhanyi, @mwoodruffdev, @TeeWeb, @ammar-s847, @AdrKacz, @phoitack, @Cy83rV37, @uranosco, @yousefokasha61

Reported