Scikit Flow is high level API that allows to create, train and use deep learning models easily with well known Scikit Learn API.

To run these examples, you need to have scikit learn library installed (sudo pip install sklearn). Some examples use the pandas library for data processing (sudo pip install pandas).

• Deep Neural Network Regression with Boston Data
• Convolutional Neural Networks with Digits Data
• Deep Neural Network Classification with Iris Data
• Grid search and Deep Neural Network Classification
• Deep Neural Network with Customized Decay Function
• Building A Custom Model
• Accessing Weights and Biases in A Custom Model
• Building A Custom Model Using Multiple GPUs
• Building A Model Using Different GPU Configurations
• Using skflow with Pipeline
• Example of Saving and Restoring Models
• Multi-output Deep Neural Network regression
• Improving Performance Using Early Stopping with Iris Data
• Out-of-core Data Classification Using Dask

• Convolutional Neural Networks on MNIST Data
• Deep Residual Networks on MNIST Data

• Text Classification Using Recurrent Neural Networks on Words (See also Simplified Version Using Built-in RNN Model using built-in parameters)
• Text Classification Using Convolutional Neural Networks on Words
• Text Classification Using Recurrent Neural Networks on Characters
• Text Classification Using Convolutional Neural Networks on Characters

• Character level language modeling

• Character level neural language translation
• Word level neural language translation