Jiangmiao Pang comments

Results 18 comments of


                                            Jiangmiao Pang

Error when using dist_train/dist_test

How many GPUs do you have on your machine? Can you try the other python versions? Like 3.7 or 3.8?

Roadmap of MMTracking

Hi @SilvioGiancola @agirbau, thanks for your interest! We are discussing a development plan in which we tend to support 2-3 new methods for each task and support the other popular...

IndexError: index 3 is out of bounds for dimension 0 with size 3 when running Tracktor with Custom Dataset

Hi, as it is your custom datasets, I cannot reproduce your bug and help you directly. Can you try to print more values in the forward process and give us...

Possibility of using pre-trained weights for neck and head and only customizing backbone ?

You should re-train the model if you modify any weights in the model.

How to track visualization on BDD

New inference code available at [here](https://github.com/SysCV/qdtrack/blob/master/tools/inference.py).

TypeError : Resnet : init() got an unexpected keyword argument 'init_cfg'

Please refer to this [doc](https://github.com/SysCV/qdtrack/blob/master/docs/INSTALL.md) and install the right version with mmcv and mmdet.

TypeError : Resnet : init() got an unexpected keyword argument 'init_cfg'

Which version of mmcv or mmdet did you use?

TypeError : Resnet : init() got an unexpected keyword argument 'init_cfg'

We develop this codebase based on mmcv 2.14.0 and mmdet 1.3.10, please update the version of your mmcv.

Backbone pre-trained on BDD100K dataset

You can get the pre-trained models from [here](https://drive.google.com/file/d/1YNAQgd8rMqqEG-fRj3VWlO4G5kdwJbxz/view).

Confusion on the relationship among samples_per_GPU, nums of GPUs and LR

Yes, that could work.