Tim Joseph

I also get only 35,54% average F-score for a 5-fold on SumMe!

Wow Any idea @KaiyangZhou ?

> Actually the model is not learning anything. I ran the pre-trained model of the theano implementation and it is giving the numbers as expected. But the funny thing is...

> pytorch implementation doesn't strictly follow the 5-fold setting, Can you elaborate?

> Yes , using their pretrained theano model only. > > https://github.com/KaiyangZhou/vsumm-reinforce/blob/master/vsum_test.py > After line 80, replace probs ( the values predicted by the model ) by adding > probs...

> The theano model reproduces the numbers shown in the paper correctly. But not using the RL model gives the same results as well . Does it also reproduce the...

When using the PyTorch implementation without folds (instead train and test on the same videos) I can get 40.1% on SumMe!

> sorry for not responding promptly, too busy with deadlines. > > typically, learning with RL is quite tricky and hard to judge how well the agent is learned unless...

Sounds good to me!

Here is an updated and isolated version: ``` from dataclasses import dataclass import hydra from clearml import Task from clearml.config import running_remotely from hydra.core.config_store import ConfigStore from omegaconf import OmegaConf...