# /path/to/reository/
> conda env create -f environment.yml
activate anaconda environment
> activate deeplearning # window
> source activate deeplearning # linux/ mac
Download dataset
deeplearning> python
>>> from keras.datasets import mnist
>>> mnist.load_data()
>>> quit()
Create new anaconda environment
> conda create -n deeplearning python=3.6
activate anaconda environment
> activate deeplearning # window
> source activate deeplearning # linux/ mac
install the library
deeplearning> conda install -c conda-forge scipy tensorflow keras pillow matplotlib jupyter
Download dataset
deeplearning> python
>>> from keras.datasets import mnist
>>> mnist.load_data()
>>> quit()
Download image file
download file from http://158.108.138.156:5000/
import dockefile
> docker load < busybox.tar.gz
> docker run --rm -p 8888:8888 wit543/deeplearning:circleci /bin/bash -c "source activate deeplearning && jupyter lab --ip='0.0.0.0' --port=8888 --no-browser --allow-root"