Sovit Ranjan Rath comments

Results 221 comments of


                                            Sovit Ranjan Rath

Faster RCNN init error

Glad to hear that.

Albumentations bouning box error

Hello, I have just pushed an update to `datasets.py`. It removes all files with invalid bounding boxes before training. I am closing the issue for now. Please re-open if needed.

precision recall curve

@EmmaVanPuyenbr Not at the moment. But I am open to contributions and PR.

training on cpu stuck in first epoch

Hi. I would not recommend training Faster RCNN on a CPU. It will take a long time. If possible, please try to use Colab or Kaggle.

precision recall curve

Carrying it over to #132

raise ValueError(f"Expected {name} for bbox {bbox} to be in the range [0.0, 1.0], got {value}.")

It looks like, for some images, the coordinate for y_min has been annotated out of the image border. You may need to check which images have that issue.

raise ValueError(f"Expected {name} for bbox {bbox} to be in the range [0.0, 1.0], got {value}.")

Hello, I have just pushed an update to `datasets.py` that removes all files with invalid bounding boxes before training. Please check. I am closing the issue for now. Please re-open...

Is there a way to generate a confusion matrix?

Hello @raulmat19 Most probably it should originate from the `evaluate()` function in the `engine.py` file. But I will have to take a closer look to implement it in a clean...

Is there a way to generate a confusion matrix?

@raulmat19 I completely understand your concern. It is a bit tricky to do with Torchvision object detection models. But I will try again for that in a few days.

Is there a way to generate a confusion matrix?

Got it. Can you post an image/example of the result that you got?