No such file or directory: ...../model_epoch_27.model'
auxvic opened this issue · comments
Hello, first of all, thank you for all the work done and making this repository public. I am trying to reproduce the steps in the readme file, first with testing and then I will try to reproduce the training. I have downloaded the data and installed the dependencies, but when running python experiments/scripts/test_tracktor.py , I get the following error:
INFO - main - Initializing object detector(s).
ERROR - test_tracktor - Failed after 0:00:01!
Traceback (most recent calls WITHOUT Sacred internals):
File "experiments/scripts/test_tracktor.py", line 93, in main
obj_detect_model, map_location=lambda storage, loc: storage)
File "/home/ubuntu/tracktor/lib/python3.6/site-packages/torch/serialization.py", line 581, in load
with _open_file_like(f, 'rb') as opened_file:
File "/home/ubuntu/tracktor/lib/python3.6/site-packages/torch/serialization.py", line 230, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/home/ubuntu/tracktor/lib/python3.6/site-packages/torch/serialization.py", line 211, in __init__
super(_open_file, self).__init__(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'output/faster_rcnn_fpn/faster_rcnn_fpn_training_mot_17/model_epoch_27.model'
I have checked that the mentioned file is inside the output folder where the repository was cloned, and also under the output folder inside the folder for the environment, since it seems there is where it tries to get the data from. But I still get the error. What am I missing? Thank you
I assume you did not download the zip file with the model files. Steo 4 in the README.
I assume you did not download the zip file with the model files. Steo 4 in the README.
I did, that's why I said I have checked the paths. Here is a screenshot.
Okay, my bad. But if you compare the paths you see they dont match. output/faster_rcnn_fpn/faster_rcnn_fpn_training_mot_17/model_epoch_27.model has another faster_rcnn_fpn subolder. I will fix this in the default cfg.
Okay, my bad. But if you compare the paths you see they dont match.
output/faster_rcnn_fpn/faster_rcnn_fpn_training_mot_17/model_epoch_27.modelhas anotherfaster_rcnn_fpnsubolder. I will fix this in the default cfg.
Oh, I'm sorry, I did not notice that intermediate folder. I will fix it manually. Thanks a lot! And also for the super quick response.
Now it starts but I get a runtime error related with torchvision and cuda. I will do some research to find it out, it is probably something related with the packages (I had to install one of them manually because the one in the requirements.txt did not work). Again, thanks!
The runtime error is most likely from a version mismatch. I hope you can figure it out!