ShevaXu / hello-tensorflow

Have fun learning TensorFlow :smile:.

Hello TensorFlow

Have fun learning TensorFlow.

Installation

Docker installation is highly recommended for experimental trial:

docker pull tensorflow/tensorflow

It is also on gcr.io/tensorflow/tensorflow besides Docker Hub.

Examples

Board

This example shows how to run a simple logistic regression and visualize it using TensorBoard:

$ docker run -it -v $PWD:/board -w /board -p 8888:8888 tensorflow/tensorflow bash

root@63410e8c4372:/board# mkdir log && python board.py
...
root@63410e8c4372:/board# tensorboard --logdir log --host 0.0.0.0 --port 8888
Starting TensorBoard 41 on port 8888
(You can navigate to http://172.17.0.2:8888)

• -p and --host 0.0.0.0 setting together make TensorBoard accessible from host's browser;
• -v (optional) maps a volume into the container so that you can modify you script on your host machine with any IDE.

Refs

• Python API (Stable since 1.0)
• TensorBoard: Visualizing Learning