oulu-imeds / deepknee Goto Github PK

Pulled the latest version of the branch pypackage and ran sh deploy.sh cpu as is. When accessing the ui-service I receive a flash message saying "Server is not connected". No errors in the terminal where I started the services.

Any suggestions?

Can you please share the link to the dataset created and made public by the authors of the paper as refered in the paper.

Hi, i have executed your code as per readme file

1. I ran detector.py (results in ROI stored in text file)
2. then i ran crop_rois_your_dataset.py (results in "5" folder)
3. then i ran predict.py and i got this result.

        6.3281e-01, 4.9861e-01, 7.8321e-01, 6.8283e-01, 5.1901e-11, 6.0996e-01,
        6.0776e-11, 8.3949e-01, 5.4392e-01, 9.4065e-01, 5.2019e-12, 2.1195e-10,
        1.6942e-11, 1.4646e-09, 1.7900e-10, 7.5311e-11, 4.3836e-11, 1.3461e-10,
        7.1184e-08, 7.2854e-01, 3.0345e-09, 1.3290e-06, 8.7507e-11, 5.1108e-11,
        6.0748e-10, 2.9482e-11, 4.5003e-09, 3.8715e-10, 4.1433e-11, 1.2130e-10,
        3.6910e-12, 7.1338e-01, 1.2946e-12, 3.6671e-11])), ('module.final.1.weight', tensor([[-5.4076e-01,  1.9848e-12,  3.5346e-01,  ...,  7.9639e-01,
         -1.0393e-30, -7.2086e-03],
        [-3.5829e-01,  4.2498e-12, -1.0540e-01,  ...,  4.3795e-01,
         -9.5742e-22, -2.1120e-03],
        [ 9.3125e-02, -6.8784e-14, -1.9588e-01,  ...,  3.0634e-01,
          3.1095e-31,  1.4625e-03],
        [ 8.6362e-01,  2.3195e-12,  4.2160e-02,  ..., -5.7820e-01,
         -2.1569e-27,  5.8158e-04],
        [ 2.4711e-02,  6.3130e-13, -1.6351e-01,  ..., -9.7592e-01,
          1.4626e-30, -6.1977e-06]])), ('module.final.1.bias', tensor([-0.1403, -0.0196, -0.2482, -0.2937,  0.6932]))])

but my text file named "KL_grading_result" is empty and also how can i see the resultant gradcam image, when i ran produce_gradcam.py i got this error

TypeError: __init__() got an unexpected keyword argument 'n_values'

can you please help me on this

E.g., from https://ieeexplore.ieee.org/abstract/document/9169719/ .

https://github.com/MIPT-Oulu/DeepKnee/blob/master/Dataset/crop_rois_your_dataset.py#L16

Including, but not limited to:

• getting rid of import magic and relative paths in https://github.com/MIPT-Oulu/DeepKnee/blob/master/inference_own/predict.py ;
• removing unused imports.

FYI, inference in branch gradcam is fully compatible with 0.4.0.

Sir, I cant handle the following error when trying to test your work.
Please kindly advice. Thanks.

(base) C:\Users\MrDig\Desktop\DeepKnee-master\own_codes>python predict.py Version of pytorch: 1.2.0 C:\Users\MrDig\Desktop\DeepKnee-master\own_codes\model.py:32: UserWarning: nn.init.kaiming_uniform is now deprecated in favor of nn.init.kaiming_uniform_. nn.init.kaiming_uniform(m.weight.data) C:\Users\MrDig\Desktop\DeepKnee-master\own_codes\model.py:36: UserWarning: nn.init.kaiming_uniform is now deprecated in favor of nn.init.kaiming_uniform_. nn.init.kaiming_uniform(m.weight.data) Traceback (most recent call last): File "predict.py", line 68, in <module> tmp = load_model(snp_name, KneeNet(args.bw, 0.2, True)) File "predict.py", line 24, in load_model state_dict = torch.load(filename, map_location=lambda storage, loc: storage) File "C:\Anaconda\lib\site-packages\torch\serialization.py", line 386, in load return _load(f, map_location, pickle_module, **pickle_load_args) File "C:\Anaconda\lib\site-packages\torch\serialization.py", line 563, in _load magic_number = pickle_module.load(f, **pickle_load_args) _pickle.UnpicklingError: invalid load key, 'v'.