Keras with tf-gpu installed but starts cpu training

0bserver07 / One-Hundred-Layers-Tiramisu

Keras Implementation of The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation by (Simon Jégou, Michal Drozdzal, David Vazquez, Adriana Romero, Yoshua Bengio)

https://arxiv.org/abs/1611.09326

MIT License

197 stars 54 forks source link

Keras with tf-gpu installed but starts cpu training #7

Closed codecolony closed 7 years ago

codecolony commented 7 years ago

Hi,

I have installed exactly as described in the instructions but the training always starts in CPU mode. I'm using tensorflow backend. Why do we need to install both tensorflow-gpu as well as tensorflow?

codecolony commented 7 years ago

solved using... export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64