https://github.com/enhuiz/digit-recognizer
Digit Recognizer - Convolutional Neural Network in Tensorflow
https://github.com/enhuiz/digit-recognizer
Last synced: 5 months ago
JSON representation
Digit Recognizer - Convolutional Neural Network in Tensorflow
- Host: GitHub
- URL: https://github.com/enhuiz/digit-recognizer
- Owner: enhuiz
- Created: 2017-06-22T11:24:18.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2017-06-22T11:35:39.000Z (over 8 years ago)
- Last Synced: 2025-02-25T04:15:20.856Z (10 months ago)
- Language: Python
- Size: 2.93 KB
- Stars: 0
- Watchers: 3
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Digit Recognizer - Convolutional Neural Network in Tensorflow
## How to run
### 1. Create a directory called data and copy your data downloaded from Kaggle into it
```
$ mkdir data
$ cp path-to-your-download-directory/train.csv data/train.csv
$ cp path-to-your-download-directory/test.csv data/test.csv
```
### 2. Install dependencies
```
$ pip3 install numpy
$ pip3 install pandas
$ pip3 install tensorflow
```
### 3. Train your model
```
python3 main.py
```
## FYI
It will take a very long time if you trying to train CNN using CPU. To enable GPU for faster training, you need a NVIDIA graphic card that supports cuda. The configurations are kind of complicated and you can find some instructions [here](https://www.tensorflow.org/install/).