the problem when i track mots20-all

[niangea]

when i use "python src/track.py with \ dataset_name=MOTS20-ALL \ obj_detect_checkpoint_file=models/mots20_train_masks/checkpoint.pth"

---

I get " ERROR - track - Failed after 0:00:26! Traceback (most recent calls WITHOUT Sacred internals): File "src/track.py", line 91, in main obj_detector.load_state_dict(obj_detect_state_dict) File "/home/lihenian/.conda/envs/bishe37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1672, in load_state_dict self.class.name, "\n\t".join(error_msgs))) RuntimeError: Error(s) in loading state_dict for DETRSegmTracking: size mismatch for class_embed.weight: copying a param with shape torch.Size([2, 256]) from checkpoint, the shape in current model is torch.Size([21, 256]). size mismatch for class_embed.bias: copying a param with shape torch.Size([2]) from checkpoint, the shape in current model is torch.Size([21]). "

this is my environment

Collecting environment information... PyTorch version: 1.13.1+cu117 Is debug build: False CUDA used to build PyTorch: 11.7 ROCM used to build PyTorch: N/A

OS: Ubuntu 18.04.5 LTS (x86_64) GCC version: (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0 Clang version: Could not collect CMake version: version 3.10.2 Libc version: glibc-2.17

Python version: 3.7.16 (default, Jan 17 2023, 22:20:44) [GCC 11.2.0] (64-bit runtime) Python platform: Linux-5.4.0-135-generic-x86_64-with-debian-buster-sid Is CUDA available: True CUDA runtime version: 11.3.109 CUDA_MODULE_LOADING set to: LAZY GPU models and configuration: GPU 0: NVIDIA GeForce RTX 3090 GPU 1: NVIDIA GeForce RTX 3090

Nvidia driver version: 515.86.01 cuDNN version: Probably one of the following: /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_adv_infer.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_adv_train.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_cnn_infer.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_cnn_train.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_ops_infer.so.8 /usr/local/cuda-11.3/targets/x86_64-linux/lib/libcudnn_ops_train.so.8 HIP runtime version: N/A MIOpen runtime version: N/A Is XNNPACK available: True

Versions of relevant libraries: [pip3] mypy-extensions==1.0.0 [pip3] numpy==1.21.6 [pip3] torch==1.13.1 [pip3] torchfile==0.1.0 [pip3] torchvision==0.14.1 [conda] numpy 1.21.6 pypi_0 pypi [conda] torch 1.13.1 pypi_0 pypi [conda] torchfile 0.1.0 pypi_0 pypi [conda] torchvision 0.14.1 pypi_0 pypi

[taxy]

Number of class heads are different between model and saved wieghts

[niangea]

Number of class heads are different between model and saved wieghts

What command should I use

[taxy]

Number of class heads are different between model and saved wieghts

What command should I use

https://github.com/timmeinhardt/trackformer/blob/e1dbc2536b86348c2f62c788c98a2faf06247147/cfgs/train_mots20.yaml#L6 I think you have a problem with the checkpoint file, the relevant config gives the answer

[niangea]

Number of class heads are different between model and saved wieghts

What command should I use

https://github.com/timmeinhardt/trackformer/blob/e1dbc2536b86348c2f62c788c98a2faf06247147/cfgs/train_mots20.yaml#L6

I think you have a problem with the checkpoint file, the relevant config gives the answer

but i can not find this model in trackformer_models_v1.zip, how can I find this model?

[taxy]

it is possible that the readme is out of date, and not all trains are available anymore. ☹️ I only tried one before using it

[niangea]

it is possible that the readme is out of date, and not all trains are available anymore. ☹️ I only tried one before using it

Have you tried it on mot17? After I tested four data on mot17, I received the error message, and then uploaded the issue

[timmeinhardt]

MOT17 is tested and works. But I have to admit that the MOTS20 models/code were not thoroughly tested after the last larger code changes. In order to make it work one has to add an exception here initializing with num_classes = 1. To see if this helps you could hardcode it for a start.