Glenn Jocher

It seems like the issue might be related to the specific configuration of your edge device. Please ensure that your device supports the required OpenVINO operations and that you have...

@LJ4309 hello! Thank you for bringing this to our attention. It looks like there might be a discrepancy between the drawn mask and the mask data returned in the results....

Renaming files to numbers can help standardize them across different systems. Make sure there are no special characters or spaces in the filenames.

Setting `model.predictor = None` at the start of a new video can help reset the tracker state, ensuring it processes each video independently.

@Murtazaabidi1 hello! For implementing knowledge distillation with YOLOv8, you can start by modifying the student model's architecture and loading the teacher model weights. To calculate the distillation loss, you can...

@Murtazaabidi1 great to hear that you've successfully trained the student model using knowledge distillation with YOLOv8! If you have any further questions or need additional assistance, feel free to ask....

Thank you for sharing your observations. Switching the roles and using the IR model as the teacher while the RGB model acts as the student could indeed help improve prediction...

It looks like the error is due to trying to subscript a `DetMetrics` object. Instead of accessing `results['dataloader']`, you should iterate over your dataset directly. Modify your training loop to...

If the issue was resolved, please share your solution here for others to benefit. For YOLOv9, you can refer to the Ultralytics documentation for guidance.

INT8 quantization in PyTorch is typically done using static quantization rather than dynamic. Dynamic quantization is more suited for models with linear layers. For full INT8 quantization, you might need...