thhung comments

Results 14 comments of


                                            thhung

can not get the satisfactory result on caltech pedestrian detection

@liuwei16 did you manage to get the same result 10.33 miss rate?

can not get the satisfactory result on caltech pedestrian detection

Hello @zhaoweicai , I am evaluating your pretrained model on the Caltech dataset for 1xtest? For different evaluation code, I only can get around 22% miss rate meanwhile, I expected...

Cannot install torch-scatter and torch-sparse

Maybe you could try this? https://stackoverflow.com/questions/21530577/fatal-error-python-h-no-such-file-or-directory

Callbacks cannot remove and add nodes

Hi, could you confirm that this issue is resolved or not?

Accept 2D motion data from webcam

interested on this one as well.

what is the gpu memory needed by the h model

I tried with h models on colab and I get this : >OutOfMemoryError: CUDA out of memory. Tried to allocate 7.40 GiB (GPU 0; 14.75 GiB total capacity; 11.25 GiB...

[Enhancement]: Use SAM in web demo for masking.

Any updates on this? I am interested on this one as well. SAM needs some interactive inputs to get the good quality mask.

run my own image error @ /examples/image_to_annotations.py

Can you guys show the messages you get from torchserve? That could help a lot to narrow down what is the cause. Try to see the log file in your...

#1155: Add support for OpenAI-compatible endpoint in LLM and Embed

Can we merge this PRs? It seems only add more config and change only the OpenAI Chat and Embeding to support the local endpoints.

Colab 无法运行

I still have the same problem after copying videos to the input folder. It seems weirds.