JagdishKolhe

Do we have any way to export to ONNX? Can we simply remove quant layers for exporting to ONNX?

### Describe the required feature NSE and BSE both do have their intraday data at EOD. Basically at EOD, we see an intraday chart we can hit the URL to...

bug in super-gradients/src/super_gradients/training/models/conversion.py

5

### Describe the bug A clear and concise description of what the bug is. ### To Reproduce Steps to reproduce the behavior: 1. net = models.get(model_name, pretrained_weights="coco") 2. models.convert_to_onnx(model=net, out_path=f"{model_name}.onnx",...

Can we have independent pytorch based script for NAS detectors (coco dataset) evaluation

5

Looks like, yolo-NAS architectures have different prediction format (different than standard coco) for model.predict(...) Can we have a distinct standalone pytorch function which does not use super-gradient code base rather...

Errors while using super_gradients.evaluate_from_recipe

2

### 🐛 Describe the bug Looks like we can evaluate the model form trained checkpoints only, evaluate_from_recipe gives error if I want to evaluate pretrained model. I also get error...

The function get_categories() can be modified to get n_objects = len(instances['categories']) So that if someone is using more than 80 classes he can use the tool, without any issues.

It will be gr8 if you can provide docker which has all things packaged, it will be very helpful for non-tech people.