showlab / UniVTG

[ICCV2023] UniVTG: Towards Unified Video-Language Temporal Grounding

Home Page:https://arxiv.org/abs/2307.16715

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Implementation on Qvhighlights

wenjiajia123 opened this issue Β· comments

Hi, I wonder if the results of qvhighlights is got by running train_mr.py? Cause I face same data format problem when it comes to eval.py. tahnk u!😊

You can use my model checkpoints with their config for inference, and you should able to reproduce the numbers.
If you want to re-train it. Please load the pretrained weight and then set the parameter settings based on the config. and you can monitor the training logs, find the one with best val scores.