/keras-digit-recognition

https://www.kaggle.com/c/digit-recognizer

Primary LanguageJupyter Notebook

Keras-Ditial-Recognition

Competition Description

MNIST ("Modified National Institute of Standards and Technology") is the de facto “hello world” dataset of computer vision. Your goal is to correctly identify digits from a dataset of tens of thousands of handwritten images.
More details please visit https://www.kaggle.com/c/digit-recognizer

Code Environment

You can use Anaconda to install the code environment in environments folder.

  1. conda env create -f kaggle_env_mac_env_mac.yml
  2. source activate kaggle
  3. jupyter notebook

Implementation Details

  1. Data processing.
  2. Data normalization.
  3. Data split into train set, dev(cross validation) set, test set.
  4. Data classification by using Deep Learning(Convolutional Neural Network) implemented in Keras.