Tutorials

Tutorial Articles

There are a few articles that can help you to start working with NeuPy. They provide a solution to different problems and explain each step of the overall process.

• Image classification, MNIST digits
• Predict prices for houses in the area of Boston

Code Examples

NeuPy is very intuitive and it’s easy to read and understand the code. To learn more about different Neural Network types you can check these code examples.

Deep Learning
Image classification - CNN	Model training: MNIST CIFAR 10 MNIST semi-supervised training with stacked autoencoders Pre-trained models: VGG16 VGG19 SqueezeNet ResNet 50 Architectures: AlexNet GoogleNet Inception v3 Visualizations: Visualizing CNN based on the pre-trained VGG19 Looking inside of the VGG19 using SOFM
Multilayer Perceptron (MLP)	Classification: MNIST IMDB review classification using CBOW and RPROP MLP MLP with categorical and numerical features Regression: Boston house price prediction Visualizations: Visualizing training process for different algorithms MLP Visualizations
Recurrent Neural Networks (RNN)	Reber Grammar Classification, GRU Shakespear text generation, LSTM
Autoencoders	MNIST, Denoising Autoencoder MNIST, Convolutional Autoencoder MNIST, Stacked Convolutional Autoencoders MNIST, Variational Autoencoder
Reinforcement Learning (RL)	Network plays CartPole game Value Iteration Networks (VIN)
Restricted Boltzmann Machine (RBM)	Feature Learning from the MNIST Images Gibbs sampling using face images
Natural Language Processing (NLP)	Classification: IMDB review classification using CBOW and RPROP MLP Reber Grammar Classification, GRU RNN Sequence to Sequence: Shakespear text generation, LSTM RNN

Competitive networks
Growing Neural Gas (GNG)	Growing Neural Gas is an algorithm that learns topological structure of the data. Code that makes animation for GNG training Making Art with GNG
Self-Organizing Feature Maps (SOFM or SOM)	Notebooks: Creating unique text style Generating Art with SOFM Visualising VGG19 using SOFM Basics: Clustering small dataset Clustering iris dataset Learning half-circle topology Comparison between different weight initialization methods Comparison between different grid types Advanced: Visualize digit images in 2D space with SOFM Embedding 30-dimensional dataset into 2D and building heatmap visualization
Linear Vector Quantization (LVQ)	Reduce number of training samples in iris dataset with LVQ3

Associative Memory
Discrete Hopfield Neural Network	Discrete Hopfield Neural Networks can memorize patterns and reconstruct them from the corrupted samples. Articles: Exhaustive explanation with example Password recovery Code: Energy function visualization Password recovery
Cerebellar Model Articulation Controller (CMAC)	Cerebellar Model Articulation Controller (CMAC) can quantize continuous space and store it inside of the memory. It's primarily used in the control systems. Sine function approximation

Radial Basis Functions (RBF)
Probabilistic Neural Network (PNN)	Music-Speech audio classification Iris dataset classification
Generalized Neural Nerwork (GRNN)	Regression using Diabetes dataset

• « Quick start

Search

Install NeuPy

pip install neupy

Learn more about NeuPy reading tutorials and documentation.

Issues and feature requests

If you find a bug or want to suggest a new feature feel free to create an issue on Github