Krishna Kanth

coding error in nextbatch

1

Line 33-34 in util.py, function next batch `if self.resize: im = np.stack([cv2.resize(im[:,:,i], (cfg.RESIZED_SIZE, cfg.RESIZED_SIZE)) for i in range(im.shape[2])], axis=2)` the image index I at line 30 is overwritten by this...

Clarification about threshold in keypoint HOTA

2

In the distance to similarity computation at https://github.com/anDoer/PoseTrack21/blob/35bd7033ec4e1a352ae39b9522df5a683f83781b/eval/posetrack21/posetrack21/trackeval/metrics/hota_pose.py#L37 I'm unable to comprehend the reason to clip the similarity to zero if the (head normalized) distance is greater than 0.5. This...

Incorrect computation of FPs in frames with zero GT annotations

2

Hi @anDoer, Thank you for adapting the HOTA for keypoint tracking. I would like to bring to your notice about possible error in computing the False positives in images where...

Incorrect arguments to SHAWrapFedex init method

1

It seems like the ```SHAWrapFedex``` missed to take ```client_cfgs```. Please check. The Line https://github.com/alibaba/FederatedScope/blob/7f0af2d7aa1537b1a033f4817fbdb62b44e10ac6/federatedscope/autotune/algos.py#L65 ```py elif init_cfg.hpo.scheduler.startswith('wrap', client_cfgs): scheduler = SHAWrapFedex(init_cfg) ``` perhaps to be changed to ```py elif init_cfg.hpo.scheduler.startswith('wrap'):...

Client-side model parameters NOT updated during the FedEx training

1

In https://github.com/alibaba/FederatedScope/blob/6dfe8d4474451c07cd7d69690a2df33a1f455f19/federatedscope/autotune/fedex/client.py#L30 the sampled hyperparams at the client are updated to ```self.trainer.cfg``` but they don't seem to take actual effect in client-side trainer. For example, the ```dropout``` in the model...

https://github.com/ShuangLI59/Diversity-Regularized-Spatiotemporal-Attention/blob/2bb350df501ebcc69edbc84529ab0206fc8a4cb3/trainers.py#L102 Line 102 of trainer.py, shouldn't the loss be summed for every image? as in only dimensions, 1 and 2 should be summed in the torch.sum(reg, dim=(1,2) and later we...

ReCogDrive-VLM-DriveLM

3

Hello Authors Can you please tell if this model is https://huggingface.co/owl10/ReCogDrive-VLM-DriveLM is specially trained on only DriveLM dataset? And what is the difference b/w this model and https://huggingface.co/owl10/ReCogDrive-VLM-8B/tree/main ? Can...