detectRecog

> You can use the `torch.onnx` package. > > When you have your trained OFA network as `ofa_network` you can sample a random subnet with `ofa_network.sample_active_subnet()` than you can cut...

> As far as I can tell, the folders are the different stages of the _progressive shrinking_ algorithm, `kernel2kernel_depth` is the training step from elastic kernel to elastic kernel and...

@zhanglianyu0911 Hey, bro. Can you give a list of configurations? After downgrading the NDK version to 15, I still get this error: Error while executing process /Users/XX/Library/Android/sdk/cmake/3.6.4111459/bin/cmake with arguments {--build...

I did not test the environment on CUDA10.2. Could you please show me the error logs?

For segmentation, you can train on pedestrians by switching class_id to 2 rather than 1 (for cars) in the dataset. The change of code for tracking should also be similar....

When test on MOTSChallenge, you should adjust the 'Cluster' module (in utils/utils.py) to support the resolution of MOTSChallenge like: ` xm = torch.linspace(0, 2, 2048).view(1, 1, -1).expand(1, 2048, 2048) ym...

I believe you did not generate the segmentation results. Before tracking, please generate the segmentation first. As we can see in 'datasets.KittiMOTSDataset.MOTSTrackCarsValOffset', at Line 198, segmentation results should be available...

Any updates @soeaver

train code is private? @MasterBin-IIAU

I attended the ECCV conference several days ago, so I missed emails from you. I feel so sorry for that. There are problems training the model and I have verified...