search

atriumlts / subpixel

subpixel: A subpixel convnet for super resolution with Tensorflow
MIT License
2.13k stars 301 forks source link

g_loss is nan for gpu #36

Closed lingz closed 7 years ago

lingz commented 7 years ago

The logs seem to show that the g_loss value is always nan for gpu, but strangely works fine on CPU. Running with CUDA 7.5, CudNN v4, GTX 1070 Tensorflow 0.10.0rc0.

lingz commented 7 years ago

Update: this seems to be an issue with versions. When running CUDA 8, CudNN v5 Tensorflow 0.12.0, it seems to work again.