ASSERT-KTH / ASSERT-KTH/VRepair

Pre-trained of best models

Open
#8 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
C
Stars
63
Forks
17
PR merge metrics
No merged PRs in 30d

Description

Hi @chenzimin ,

The models that you provided (~270GB) is currently too large for me to handle right now and using my current resource to re-train the best model according to your report is not possible for me right now either(I had to reduce the attention heads to 3 for the model to train without the out-of-memory error). From what I understand, the models that you provided in the onedrive is all the models that you trained during the ablation study so is it possible for you to upload only your best models (preferably somewhere I can `wget` my linux server cause using onedrive force me to first download it to my local then reupload to the server) ?
Thank you a lot.

Best regards,
Dang

Contributor guide

No contributing guide indexed for this repository

Research direction

No source file, test, or entry point is mentioned. First clarify which models from the ablation study are considered the best and where they should be hosted for wget access. Done means the selected pretrained models are available for download without requiring the full ~270GB collection.

Written by the indexing model from the issue text.

Assessment

Tech stack
c, machine-learning
Domain
machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.