Alan Bi

It looks like your model might be overfitted or basically just producing too many false positives. What does your training dataset look like? It might be worth going back and...

Maybe something like maplotlib's [savefig](https://matplotlib.org/api/_as_gen/matplotlib.pyplot.savefig.html#matplotlib.pyplot.savefig) could be useful for this

Would you be able to share the code that causes this issue? So is the output file missing completely or generated but has no frames?

Could you check to make sure all the paths to files are correct? If a file is being written but is empty, then that probably means there's an issue when...

Not sure - it's hard to say anything without having more information about the scenario. However, you said that the predictions are working fine?

In that case it's possible it's not a bug with the loss values - feel free to take a look at the [torchvision documentation](https://pytorch.org/vision/stable/models.html#object-detection-instance-segmentation-and-person-keypoint-detection) to learn about what the loss...

How many images are in the validation dataset? If it's a small number then it's possible it's looping through all of them, but the bar just doesn't show the updated...

It looks like this might be the case by default with PyTorch's models. I'd recommend taking a look at [this thread](https://discuss.pytorch.org/t/only-100-predictions-outputs-for-torchvision-object-detection-finetuning-tutorial/98029) and seeing if you can manually override that option...

Currently there's no official way to do this, but if you wanted to, you can use the `xml_to_csv` function to generate a pandas dataframe, and then only select the rows...

Be sure you're using the latest version: https://github.com/alankbi/detecto/releases/tag/v1.2.1