https://github.com/itdxer/kaggle-cooking
A very simple solution to the Kaggle's "What's Cooking?" competition :egg:
https://github.com/itdxer/kaggle-cooking
Last synced: 3 months ago
JSON representation
A very simple solution to the Kaggle's "What's Cooking?" competition :egg:
- Host: GitHub
- URL: https://github.com/itdxer/kaggle-cooking
- Owner: itdxer
- Created: 2016-04-18T19:20:38.000Z (about 9 years ago)
- Default Branch: master
- Last Pushed: 2016-04-18T19:30:58.000Z (about 9 years ago)
- Last Synced: 2023-08-03T03:55:17.322Z (almost 2 years ago)
- Language: Python
- Homepage:
- Size: 2.13 MB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# The Kaggle's "What's Cooking?" competition.
The main information about the competition you can find [here](https://www.kaggle.com/c/whats-cooking)
## Instruction on how to use
* Install Python's requirements
```bash
$ pip install -r requirements.txt
```
* Train classifier
```bash
$ python train_model.py
```
* Create submission file
```bash
$ python train_model.py --mode submission
```
The main submission file you can find in the `data` directory.
To make your result reproducible you need to add ``--reproducible`` flag
```bash
$ python train_model.py --mode submission --reproducible
```
## Tests
To run unit tests you need to execute the following command in the terminal from project's root directory
```bash
$ nosetests
```
## Documentation
```
$ python train_model.py -h
usage: train_model.py [-h] [-m {validation,submission}] [-r]
optional arguments:
-h, --help show this help message and exit
-m {validation,submission}, --mode {validation,submission}
set up training mode
-r, --reproducible make training reproducible
```