My solution for "Zalando's Fashion MNIST" challenge on dockerized Jupyter Notebook.
Before starting, download dataset to your work directory, extract it, and make sure you have fashion-mnist_test.csv
and test fashion-mnist_train.csv
files in project directory.
Dataset download link: https://www.kaggle.com/zalando-research/fashionmnist/downloads/fashionmnist.zip/4
Note: You have to be logged on Kaggle to download dataset.
.
├── zalando-fashion-mnist
| ├── fashion-mnist_test.csv
| ├── fashion-mnist_train.csv
| ├── build.sh
| ├── .gitignore
| ├── Dockerfile
| ├── LICENSE
| ├── README.md
| ├── helpers.py
| ├── main.py
| ├── main.ipynb
$ git clone https://github.com/0x01h/zalando-fashion-mnist.git
$ cd zalando-fashion-mnist
$ sh build.sh
- Copy Jupyter Notebook session token that you are able see it in terminal.
- Go http://127.0.0.1:8888 from your browser and sign in using token.
- Open work directory.
- Run main.ipynb.