- Neural Networks
- Convolutional Neural Networks
- Recurrent Neural Networks
- Generative Adversarial Networks
Project 1: Your first neural network
In this project we had to build a neural network from scratch and apply it on a real problem: predict the number of future bike rentals after analyzing a dataset containing information about a period of two years of rentals.
Project 2: Image Classification
In this project we had to build a convolutional neural network using TensorFlow and use it to classify images from the CIFAR-10 dataset.
Project 3: Generate TV Scripts
In this project we had to build a Recurrent Neural Network (RNN) using TensorFlow and use it to generate a new Simpsons TV script for a scene at Moe's Tavern using part of the Simpsons dataset of scripts from 27 seasons.
Project 4: Translate a Language
In this project we had to build a Sequence to Sequence model that can translate new sentences from English to French and train it on a dataset of English and French sentences.
Project 5: Generate Faces
In this project we had to build a Generative Adversarial Network that can generate novel faces when trained on the CelebFaces Attributes Dataset (CelebA).
Exercises with topics not covered on the projects
Lab 1: Autoencoders
In this exercise we had to implement autoencoder models and use them on examples of data compression and image denoising.
Lab 2: Transfer Learning
In this exercise we had to build an image classifier RNN using the transfer learning technique with VGGNet trained on the ImageNet dataset to classify images of flowers.
Click here to see the certificate