Chao Li

I met a similiar issue, and solved it by setting pin_memory=false. https://discuss.pytorch.org/t/using-pined-memory-causes-out-of-memory-error-even-though-batch-size-is-set-to-low-values/30602

I also got a similar prediction, all of the output is black. The minibatch loss is -1.00. Does anyone know what's wrong with it?

> Tried to make dice_coefficient loss function range from 0 to 1, by writing loss = 1.0 - (2 * intersection / (union)), but still end up same. > When...

I am also confused about the number of categories in patch_clf_train.py, It seems that the number should be 5 according to the paper. Have you run through the patch_clf_train? @HbueSky

Could you please provide the source code of converting mxnet to caffe model?

I met the same problem, I used mobilenet as backbone, and the training loss becomes Nan after two epoch. I had reduced the initial learning rate to 0.001.

I also meet the same issue. I get some images from the Visualizations of sorting all 2.37B images from LAION 5B (http://captions.christoph-schuhmann.de/aesthetic_viz_laion_sac+logos+ava1-l14-linearMSE-en-2.37B.html). I used the sac+logos+ava1-l14-linearMSE.pth to predict the scores...

In the test stage, the final score of an image patch is the formulation as above: S = ω × SDeepDet + (1 − ω) × SDeepVer. While in the...