About model COVID-Net CT-2 L (2A)

[zhourui0618]

Hi,
thanks for sharing your great work,
When i run a code in python,i get something in trouble:
2021-06-09 19:36:44.869132: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cublas64_100.dll
2021-06-09 19:37:46.729415: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library cudnn64_7.dll
2021-06-09 19:49:41.190614: W tensorflow/stream_executor/cuda/redzone_allocator.cc:312] Internal: Invoking ptxas not supported on Windows
Relying on driver to perform ptx compilation. This message will be only logged once.
confusion matrix:
[[12245 0 0]
[ 7395 0 0]
[ 5981 29 8]]
COVID-19 PPV: 1.0
COVID-19 sensitivity: 0.0013293452974410102
Normal PPV: 0.4779282619726006
Normal sensitivity: 1.0
Pneumonia PPV: 0.0
Pneumonia sensitivity: 0.0
accuracy: 0.4775508613297997
I use model "COVID-Net CT-2 L (2A)" to run my code,and I seem to get a wrong result.
Do you know why the model is not useful in my python program?

Thanks
Stay healthy!

[haydengunraj]

This seems odd to me, but I'll need more information to get to the bottom of it.

First, are you using your own data or the provided test data? If you're using your own data, is it public data that I can access?

Second, are you using custom code or the provided testing script? If you're using custom code, is there any way I can have a look at it?