Deep Learning with Tensorflow 2.0

A small collection of very basic learning tasks implemented with TF 2.0.

Feel free to file an issue if you'd like to see new tasks added, or send pull requests to contribute.

Vision

• MNIST: Simple feed-forward network
• CatvDog: Model with 4 convolutional-maxpooling blocks + data augmentation
• CatvDog: Using VGG pre-trained model (feature extraction)
• CatvDog: Transfer learning (fine-tuning) using VGG pre-trained model (to be converted to notebook)

NLP

• Reuters Multiclass: FF netowrk with one-hot encoding
• IMDB Sentiment: FF network with one-hot encoding
• IMDB Sentiment: FF network with embedding layer
• IMDB Sentiment: FF network using GloVe pre-trained embedding
• IMDB Sentiment: LSTM model

Regression

• Boston Housing Prices: Simple FF model
• Temperature Prediction: Bidirectional GRU model

Misc

• Deep Dream
• Gatys Style Transfer
• Variational Autoencoder
• LSTM Text Generation
• Transfer Learning

Tensorflow Tutorials

• A whirlwind introduction to Tensorflow 2.0