fyangf

@weichen582 who did the experiments can share some info.

I'm getting 77.8 for L1 from my own implementation without distillation. By tuning the hyperparameters I'm expecting 78+ but it is still far from 79.2 as reported in the paper....

Thanks for the clarification! That makes more sense now.

Pinged @yuanliangzhe offline

Could you paste the whole trace for debugging so the author can take a look?

We don't support YoloV5 in official models.

> @fyangf To confirm, you'd like me to move the changes from > > `official/vision/beta/projects/yolo/README.md` => `official/projects/yolo/README.md` > > and > > `official/vision/beta/projects/yolo/configs/experiments/yolov4/detection/yolov4_tiny_416_tpu.yaml` => `official/projects/yolo/configs/experiments/yolov4/detection/yolov4_tiny_416_tpu.yaml`? Yes exactly.

> @fyangf Is there anything you need on my end? Can you sync to master branch again and resolve the conflict? There are some new changes to the README file.

These packages are needed for the entire library as it contains both nlp and vision, plus many advanced training and optimization techniques. If you have strict security requirement and only...

Object Detection API is now maintained by the community so you are welcome to contribute to this feature. If you are talking about the models under `official`, then we do...