lucasxlu / comboloss Goto Github PK

ComboNet (SEResNeXt50) | CombinedLoss (alpha=2, beta=1, gamma=1) | 0.2191 | 0.2891 | 0.9066
ComboNet (SEResNeXt50) | CombinedLoss (alpha=2, beta=1, gamma=1) | 0.2126 | 0.2813 | 0.9117

same config, two different result
because of randomness?

hi lucasxlu
Thank you very much for posting the code on GitHub
I used your code in Docker but I got an error while loading the model
Can you guide me on this
The error text is as follows

Traceback (most recent call last):
web_1 | File "/usr/local/lib/python3.7/site-packages/django/apps/config.py", line 244, in create
web_1 | app_module = import_module(app_name)
web_1 | File "/usr/local/lib/python3.7/importlib/init.py", line 127, in import_module
web_1 | return _bootstrap._gcd_import(name[level:], package, level)
web_1 | File "", line 1006, in _gcd_import
web_1 | File "", line 983, in _find_and_load
web_1 | File "", line 953, in _find_and_load_unlocked
web_1 | File "", line 219, in _call_with_frames_removed
web_1 | File "", line 1006, in _gcd_import
web_1 | File "", line 983, in _find_and_load
web_1 | File "", line 965, in _find_and_load_unlocked
web_1 | ModuleNotFoundError: No module named 'classification_branch'
web_1 |
web_1 | During handling of the above exception, another exception occurred:
web_1 |
web_1 | Traceback (most recent call last):
web_1 | File "manage.py", line 22, in
web_1 | main()
web_1 | File "manage.py", line 18, in main
web_1 | execute_from_command_line(sys.argv)
web_1 | File "/usr/local/lib/python3.7/site-packages/django/core/management/init.py", line 419, in execute_from_command_line
web_1 | utility.execute()
web_1 | File "/usr/local/lib/python3.7/site-packages/django/core/management/init.py", line 395, in execute
web_1 | django.setup()
web_1 | File "/usr/local/lib/python3.7/site-packages/django/init.py", line 24, in setup
web_1 | apps.populate(settings.INSTALLED_APPS)
web_1 | File "/usr/local/lib/python3.7/site-packages/django/apps/registry.py", line 91, in populate
web_1 | app_config = AppConfig.create(entry)
web_1 | File "/usr/local/lib/python3.7/site-packages/django/apps/config.py", line 250, in create
web_1 | app_config_class.qualname,
web_1 | django.core.exceptions.ImproperlyConfigured: Cannot import 'classification_branch.bias'. Check that 'process.apps.ProcessConfig.name' is correct.

Hi, I got this warning

NVIDIA GeForce RTX 3080 Laptop GPU with CUDA capability sm_86 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.
If you want to use the NVIDIA GeForce RTX 3080 Laptop GPU GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/

My compute capability: 8.6

Fix to this:

python3.7 -m venv .venv
source .venv/bin/activate
pip3 install torch==1.9.0+cu111 torchvision==0.10.0+cu111 torchaudio==0.9.0 -f https://download.pytorch.org/whl/torch_stable.html

Validate that it fixed:

$ python3
Python 3.7.12 (default, Mar 11 2022, 12:03:27) 
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
t>>> torch.cuda.get_arch_list()
['sm_37', 'sm_50', 'sm_60', 'sm_70', 'sm_75', 'sm_80', 'sm_86']
>>> exit()

Hope this helps you!

Hi,
thanks for sharing your work.
In your code, I can find the training code and the pre-trained model, but I wonder if you can kindly share the test code that uses your pre-trained model.

Thanks in advance. BR.