TensorFlow/Keras cheat sheet

This repository provides cheat sheets for building neural networks using TensorFlow/Keras.

Medium article

See the article on Towards Data Science.

Contents

The cheat sheet provides the data about following topics:

• Typical Neural Network Architectures with Sequential API;
• More Complex Neural Network Architectures with Functional API;
• Compile the model;
• Train the model;
• Explore learning curves;
• Evaluate the model;
• Save and load the model.

And also about:

• Using ImageDataGenerator;
• Tokenizing and padding sentences for NLP tasks.

Repository Content

• tf_cheat_sheet.ipynb contains a cheat sheet in a Jupyter notebook
• tf_cheat_sheet.py contains cheat sheet in a Python script
• special_variable_example.ipynb file contains an example of using a special _ variable in Jupyter Notebook / Google Colab notebooks