100+ Keras Coding Exercises for AI Programming

Learn the basics of the Keras Python library, its setup, and how to build, train, and evaluate a simple neural network using sequential and functional APIs.

Complete six coding exercises with the Keras API to explore neural networks and deep learning, build and train models, and compare sequential and functional approaches.

Learn to work with data in Keras through coding exercises, covering loading and managing datasets, normalization and one-hot encoding, train/validation/test splits, data augmentation, and handling missing data and scaling features.

Explore building neural networks with Keras through six coding exercises and a video lecture, covering dense layers, activation functions, loss functions, optimizers, dropout, and batch normalization.

This course offers Python Keras coding exercises with algorithmic solutions, teaching you to build neural networks with dense layers, activation functions, optimizers, dropout, and batch normalization.

Learn to train and evaluate Keras models by compiling with optimizers and metrics, training with fit, evaluating performance, and using early stopping and checkpointing to tune learning rate.

Practice coding exercises that explore compiling a Keras neural network with different optimizers and metrics. Train, evaluate, visualize training history, and experiment with hyperparameters, early stopping, and model checkpoint.

Explore convolutional neural networks in Keras through practical coding exercises, building a basic CNN for image classification, and applying filters, pooling, dropout, and batch normalization.

Develop and train a basic cnn for image classification in python using keras, then extend with more layers, visualize filters, and explore transfer learning with VGG or ResNet.

Learn recurrent neural networks and LSTMs using Keras, implement simple RNNs, LSTMs, and GRUs, apply to text data such as sentiment analysis, and build sequence-to-sequence models for time series forecasting.

Practice six hands-on Keras coding exercises, including sentiment analysis, covering sequential data with RNN, LSTM, GRU, and sequence-to-sequence time series forecasting, with data pre-processing, model training, and evaluation.

Master custom layers and activation functions in Keras, build complex models with the functional API, and complete six coding exercises with solutions.

Practice Keras programming with coding exercises on custom layers and activation function, and apply callbacks with early stopping and best-model saving, lambda layers, custom loss functions, matrices, and multi-input/output models.

Explore generative models with Keras through six coding exercises on autoencoders, variational autoencoders, and generative adversarial networks for image compression and generation.

Explore six coding exercises on generative models with Keras, including autoencoders, variational autoencoders, and generative adversarial networks, to compress, generate, and fine-tune image data.