This is a project developed using 'Build a TensorFlow Image Classifier in 5 Min' Youtube video, and this Google CodeLab as a guide.

Anyone can use this classifier to classify a cloud in Cumulus, Cumulonimbus or Cirrocumulos.

• Docker
• Python 2.7

Clone this repo to ${HOME}/tf_files/

Start the docker image

docker run -it \
  --publish 6006:6006 \
  --volume ${HOME}/tf_files:/tf_files \
  --workdir /tf_files \
  tensorflow/tensorflow:1.1.0 bash

Then, run the classifier.py script to classify the image. python classifier.py <path_to_file>

It will display a list of the trained clouds labels with their matching score accuracy.