Debugging Breakpoints Not Triggering in Model Files During Fine-tuning/Inference

[gray-husky]

Search before asking

• [x] I have searched the RF-DETR issues and found no similar feature requests.

Description

Hello, I'm experiencing an issue while trying to debug the RF-DETR model architecture. I've set breakpoints in several key files to understand the network design better, but they don't seem to be triggered during fine-tuning or inference.

Files where I've placed breakpoints: models/lwdetr.py models/backbone/projector.py models/backbone/backbone.py

Problem: None of these breakpoints are being hit during execution, which makes me wonder if these files are actually being used during the fine-tuning or inference process. I'd like to confirm whether these components are indeed part of the execution path.

Questions: 1.Are these files (lwdetr.py, projector.py, backbone.py) actually executed during standard fine-tuning or inference? 2.If not, could you please guide me to the correct files where the main model architecture is implemented?

Use case

No response

Additional

No response

Are you willing to submit a PR?

• [ ] Yes I'd like to help by submitting a PR!

[isaacrob-roboflow]

Those files are used as part of the network definition :) they're not directly 'executed' but they ARE imported from