Sovit Ranjan Rath

If you can create a PR that would be great. I pushed some changes yesterday. But they were to `train.py` and `datasets.py`. Basically, mosaic augmentation was applied all the time...

Yes, it is possible, but right now `eval.py` does not have an argument parser to control it from the command line. You can use the `iou_thresholds` argument while initializing `MeanAveragePrecision`...

Hello @utility-aagrawal It will load the COCO pretrained weights by default. You only need to provide --weights if you want to continue from one of your checkpoints.

Hi. Try using `--imgsz 640` along with square resizing and AMP (Automatic Mixed Precision). Along with your command here are the additional arguments. Using AMP you can give double the...

Sure. Thanks.

Hmm... That can be because of the fasterrcnn_resnet50_fpn_v2 model. Did you try with fasterrcnn_resnet50_fpn model?

Faster RCNN is certainly slow to train compared to YOLO. However, it is not because of the data loader. Instead its because of the two stage nature of Faster RCNN.

Hi, please do `pip install -r requirements.txt`

Hi. Can you train again? I have pushed an update where you can export at 640x640 and run inference at any resolution. Basically dynamic export. While exporting do not give...

Hi. The COCO dataset YAML file in the `data_configs` directory expects 81 classes instead of 91 classes. You can find the correct dataset [here on Kaggle](https://www.kaggle.com/datasets/sovitrath/coco-xml-format). If you are facing...