darkon
darkon copied to clipboard
Toolkit to Hack Your Deep Learning Models
Hello, nice work, i really appreciate it. I would like to know if it's possible to apply the idea of gradCam to a network such: pre-trained CNN for feature extraction...
test_batch() function required in InfluenceFeeder for large test data feeding
I tired to use darkon influence to find the influential example for wrong predicted examples. However, one example took about 45 mins to complete. Is it that slow?
There are many pre-trained nets in Slim, can darkon to visualize other nets, like inception?
Using * `opencv` for image processing * `scikit-image` for 1-D interpolation
Hi, I came across your implementation of the Influence Function and am interested in separately grabbing the Hessian that is needed for the calculations. Is there a way for me...
Implement parallel processing of _grad_diffs() and _grad_diffs_all for faster execution Needs to modify Tensorflow gradient functions