superaha

NMS free version only achieves 41.7 mAP

4

I try to implement the NMS-free version of YOLOX, as reported in the paper, NMS free only do little harm to the mAP performance. My current implementation changes the `self.dynamic_k_matching`...

Performance drops significantly when switching from apex to torch.cuda.amp

8

Hi there, As discussed here: [apex vs torch.cuda.amp](https://discuss.pytorch.org/t/torch-cuda-amp-vs-nvidia-apex/74994). Apex is not encouraged to use anymore. I have implemented an version, which replaces apex with torch.cuda.amp. However, after 300 epochs, the...

Table 4 and Table 5 using COCO pretraining or not?

4

Hi there, Thank you for sharing the repo. In **Table 3**, the results of YOUTUBE-VIS 2019 are reported using both models with and without the COCO pretraining. How about **Table...

bar_mu computation is different from the paper in Eq (5)

2

Hi there, It seems the bar_mu computation is different. Should be multiplying a "-1". (below Eqn (5), bar_mu = (gt - mu_pred) / sigma. As shown here: https://github.com/Jeff-sjtu/res-loglikelihood-regression/blob/203dc3195ee5a11ed6f47c066ffdb83247511359/rlepose/models/regression_nf.py#L134 This does...

[BUG] zero2 and zero3 has different behavior using the same hyperparameter to train a large model

24

**Describe the bug** Zero3 and zero2 grad_norm are totally not in the same scale, which eventually leads to the crash of zero2's loss. **Expected behavior** Zero2 should have the same...