Public Repository

Last pushed: 5 months ago
Short Description
Python + Jupyter + Theano + TensorFlow + Keras + CUDA
Full Description

Python + CUDA 6.5

  • Python 2.7.6
  • CUDA 6.5 (nvidia driver 340.93)
  • jupyter (4.2.0)
  • theano (0.8.2)
  • tensorflow (0.10.0)
  • numpy (1.11.1)
  • scipy (0.18.1)
  • keras (1.0.8)
The following command starts a container with the Notebook server listening for HTTP connections on port 8888 with authentication configured.
docker run -d -it -p 8888:8888 --name=<name> -h <nodename> -e PASSWORD="YOURPASS" --device /dev/nvidiactl --device /dev/nvidia-uvm --device /dev/nvidia0 --device /dev/nvidia1 --volume=/<shared-dir>/:/root/shared/ tblab/theano-cu6.5-tf
Testing in batch mode

docker exec -it <name> /root/samples/deviceQuery
docker exec -it <name> python /root/samples/code/gpu2.py
docker exec -it <name> python -m tensorflow.models.image.mnist.convolutional

Jupyter Notebook

http://jupyter.org/

Docker Pull Command
Owner
tblab

Comments (0)