aveysov

Hi @nerox8664, Many thanks for your effort and support! Pulled your repo anew. **Did the following tests:** - plain inceptionv4 w/o imagenet pre-training (your tests/inception.py) - gave **less than 0.0**,...

Also, this just may be due to the fact that my model is just binary classification, and for random data it is very confident in one class...

Check out my solution here - https://github.com/amdegroot/ssd.pytorch/issues/72