Deep Learning Showcase

1 minute read

Keras Use Cases

Various Python scripts using Keras artificial neural networks to solve deep learning problems, such as Image Classification and LSTM Text Generation.

Neural Networks

Binary Classification with Keras

Implements a movie review classifier using the IMDB dataset that comes packaged with Keras.
Click here for Jupyter Notebook

Multi-class Classification with Keras

Implements a news classifier using the Reuters dataset that comes packaged with Keras.
Click here for Jupyter Notebook

Regression with Keras

Implements a housing price regression model using the Boston Housing Price dataset that comes packaged with Keras.
Click here for Jupyter Notebook

Image Classification

Convnet with Keras - Digit Images

Implements a ConvNet model that classifies images in the MNIST digit dataset that comes packaged with Keras.
Click here for Jupyter Notebook

Convnet with Keras - Color Images

Implements a ConvNet model that classifies images in the CIFAR10 small images dataset that comes packaged with Keras. This is a dataset of 50,000 32x32 color training images and 10,000 test images, labeled over 10 categories.
Click here for Jupyter Notebook

Resnet with Keras

Performs images classification using Keras’ ResNet50 model.
Click here for Jupyter Notebook

Recurrent Neural Networks

Text Processing

Transforms text input into tokens and convert those tokens into numeric vectors using one-hot encoding and feature hashing. Builds basic text-processing models using recurrent neural networks (RNN). Demonstrates how word embeddings such as Word2Vec can help improve the performance of text-processing models.
Click here for Jupyter Notebook

References:

These files contain code from Deep Learning with Python, www.manning.com/books/deep-learning-with-python, Copyright 2018 Francois Chollet

Click here to access Keras Source Code Folder

Updated: