MLNSC_2017A

Public Repository of Machine Learning Nagoya Sub-Committee (MLNSC) DLScratch-Reading and Workshop.

Official Docker Image

MLNSC official Docker Image is available ( https://hub.docker.com/r/antimon2/mlnsc-dlscratch ).
Easy to use with the provided scripts.

CLI

for Linux/macOS:

$ docker pull antimon2/mlnsc-dlscratch
$ ./docker.run.sh

for Windows (PowerShell):

PS> docker pull antimon2/mlnsc-dlscratch
PS> ./docker.run.bat

Run Jupyter notebook

$ ./docker.run.jupyter.sh

PS> ./docker.run.jupyter.bat

and then open http://localhost:8888/

Setup `dataset` directory

Since Chapter 3, the workshop requires some dataset provided by official repository. Please run the script as follows.

for Linux/macOS

Run the script as below:

$ cd dataset
$ ./get_mnist_py.sh

or, if you caught some errors when runnning this, you run docker-container first and run the script as follows:

$ ./docker.run.sh
root@xxxxxxxxxxxx:/opt/notebooks# cd mlnsc/dataset
root@xxxxxxxxxxxx:/opt/notebooks/mlnsc/dataset# ./get_mnist_py.sh
root@xxxxxxxxxxxx:/opt/notebooks/mlnsc/dataset# exit

for Windows

You run docker-container first with ./docker.run.bat, and run the script in the container as follows:

root@xxxxxxxxxxxx:/opt/notebooks# cd mlnsc/dataset
root@xxxxxxxxxxxx:/opt/notebooks/mlnsc/dataset# ./get_mnist_py.sh
root@xxxxxxxxxxxx:/opt/notebooks/mlnsc/dataset# exit

antimon2/MLNSC_2017A