Skip to content

Explored deep learning of CNNs with notMNIST, and of Word2Vec and LSTM with Text8.

License

Notifications You must be signed in to change notification settings

hangyao/Udacity_deep_learning

Repository files navigation

Deep Learning

  • 1_notmnist: Preprocess notMNIST data and train a simple logistic regression model on it.
  • 2_fullyconnected: Train a fully-connected network using Gradient Descent and Stochastic Gradient Descent.
  • 3_regularization: Use regularization techniques to improve a deep learning model.
  • 4_convolutions: Design and train a Convolutional Neural Network.
  • 5_word2vec: Train a skip-gram model on Text8 data and visualize the output.
  • 6_lstm: Train a Long Short-Term Memory network to predict character sequences.

Install

This project requires Python 2.7 and the following Python libraries installed:

You will also need to have software installed to run and execute an iPython Notebook

Code

Template code is provided in the following notebook files:

  • 1_notmnist.ipynb
  • 2_fullyconnected.ipynb
  • 3_regularization.ipynb
  • 4_convolutions.ipynb
  • 5_word2vec.ipynb
  • 6_lstm.ipynb

Run

In a terminal or command window, navigate to the top-level project directory notMNIST/ (that contains this README) and run one of the following commands:

ipython notebook 1_notmnist.ipynb 
jupyter notebook 1_notmnist.ipynb

This will open the iPython Notebook software and project file in your browser.

Data

Notebook #1 to #4 use the notMNIST dataset to be used with python experiments. This dataset is designed to look like the classic MNIST dataset, while looking a little more like real data: it's a harder task, and the data is a lot less 'clean' than MNIST.

Notebook #5 uses the Text8 dataset.

License

The contents of this repository are covered under the MIT License.

About

Explored deep learning of CNNs with notMNIST, and of Word2Vec and LSTM with Text8.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published