Wonderful work! I want to know how many GPUs did you use to train on that pipeline？

The specific issues are as follows:
Traceback (most recent call last):
File "tools/test.py", line 181, in
main()
File "tools/test.py", line 87, in main
if args.work_dir is not None:
AttributeError: 'Namespace' object has no attribute 'work_dir'
Traceback (most recent call last):
File "tools/test.py", line 181, in
main()
File "tools/test.py", line 87, in main
if args.work_dir is not None:
AttributeError: 'Namespace' object has no attribute 'work_dir'
ERROR:torch.distributed.elastic.multiprocessing.api:failed (exitcode: 1) local_rank: 0 (pid: 1554819) of binary: /data1/anaconda3/envs/OVTrack/bin/python
Traceback (most recent call last):
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/run.py", line 723, in
main()
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/elastic/multiprocessing/errors/init.py", line 345, in wrapper
return f(*args, **kwargs)
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/run.py", line 719, in main
run(args)
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/run.py", line 710, in run
elastic_launch(
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 131, in call
return launch_agent(self._config, self._entrypoint, list(args))
File "/data1/anaconda3/envs/OVTrack/lib/python3.8/site-packages/torch/distributed/launcher/api.py", line 259, in launch_agent
raise ChildFailedError(
torch.distributed.elastic.multiprocessing.errors.ChildFailedError:

tools/test.py FAILED

Failures:
[1]:
time : 2024-04-28_16:25:13
host : ubuntu
rank : 1 (local_rank: 1)
exitcode : 1 (pid: 1554820)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html

Root Cause (first observed failure):
[0]:
time : 2024-04-28_16:25:13
host : ubuntu
rank : 0 (local_rank: 0)
exitcode : 1 (pid: 1554819)
error_file: <N/A>
traceback : To enable traceback see: https://pytorch.org/docs/stable/elastic/errors.html
I hope to receive your answer！

Thanks for your wonderful jobs! I followed the “GET_STARTED.md” to test the TAO model on the Open-vocabulary MOT benchmark, however, the code threw an error.

I noticed in your paper that you mentioned, 'In the validation set of TAO, 295 of the overlapping classes are present. 35 of these classes are defined as rare, which serve as our Cnovel.' and 'In the test set of TAO, there are 324 of the overlapping classes mentioned above present. 33 of these classes are defined as rare, which serve as our Cnovel.' I'd like to inquire whether the novel classes set in your benchmark (TAO's novel classes) overlap with LVIS's novel classes?

If I want to replicate the training step of CLIP distillation, how should I proceed? Can it be achieved through the code in this repository? For example, setting the freeze_detector parameter to False. I feel like this might not be sufficient.
I am very curious about how you obtained weight files such as ovtrack_clip_distillation.pth. Did you use DetPro's training code to obtain the weights for text_head and image_head?
Thanks in advance.

ModuleNotFoundError: No module named 'tao'

Can you share the code for testing Reginclip with Tracktor on the tao dataset

I followed the “GET_STARTED.md” to Train the tracker, however, the code threw an error.

why this happend?

Table 1 in the article shows the TETA result validation set and test set of OVTrack. The TAO evaluation page on motchallenge.net only provides the Track mAP metric. Where can we obtain the TETA result?

Thanks!😁😁

For running the model on an arbitrary video the GETTING_STARTED instructions suggest to use ovtrack_vanilla_clip_prompt.pth checkpoint. Where can I get it from?

Hello,
I noticed that the training cost is not mentioned in the main paper as well as in the supplementary material. I would like to know the GPU model and the number of GPUs required to train the model, as well as the training time. Also, approximately how long will the training script be released?
Thank you.

Thanks for your work!
It seems that there's no instruction to generate lvis_classes_v1.txt. Could you please tell me how to generate this, or just share this file.
Thank you in advance.

Hello, when the full training code will be released, as you said six months ago?