Wentao Bao

These five files are scripts to train different models based on the i3d backbone. In README, I simply used the `edlnokl_avuc_debias` model as an example. You definitely could try others.

No, you should know which model you want to train.

In the directory `experiments/i3d`, you could try to remove/rename the folder `results` first, then run the command `bash run_get_threshold.sh 0 edlnokl_avuc_debias 2`. If the error still appears, you may need...

line 66 is the checkpoint file of trained model, and line 70 is the file to store results by`run_ood_detection.sh` . The `ValueError` should not be related to these lines. You...

I just tested the command `bash run_ood_detection.sh 0 HMDB edlnokl_avuc_debias`, which should be the same as yours. But I didn't find any error reported. Actually, the `debias_head` is not used...

@simbaforrest @RoboticsYimingLi Thank you all! I've seen the license now.

Could anyone provide a substitution for the deprecated `torch.spmm`?

You should check your annotation files carefully and make sure your class names are exactly the same as the configuration in VOCcode/VOCinit.m. And make sure ./imdb/cache/ folder clean before your...

Hi Mengyuan, Thanks for the help! Here I uploaded the training and evaluation splits to the [Google Drive](https://drive.google.com/drive/folders/1iR24CssAfpD21h7EmX8JLpPn17XY-Rfo?usp=sharing).

I also get the similar results, that the average precision is 69.17%, lower than 73.53% reported in the paper. I also noticed that the maximum epoch in this code is...