TensorFlow 2 for Deep Learning Specialization.

Course - 1

Getting started with TensorFlow 2 - Coursera

Table of Contents

• [Week 1]
  • Lesson Topic: Introduction to TensorFlow, TensorFlow 2, TensorFlow in Google Colab
• [Week 2]
  • Lesson Topic: The Sequential model API, Feedforward neural networks, Convolutional neural networks, Weight initialization, Compiling your model, Training your model, Evaluation and prediction
  • Assignment: CNN classifier for the MNIST dataset
• [Week 3]
  • Lesson Topic: Validation, regularization and callbacks, Model validation, Model regularization, Batch normalisation, Callbacks, The logs dictionary, Early stopping and patience, Additional callbacks
  • Assignment: Model validation on the Iris dataset
• [Week 4]
  • Lesson Topic: Saving and loading model weights, Explanation of saved files, Model saving criteria, Saving the entire model, Saving model architecture only, Loading pre-trained Keras models, TensorFlow Hub modules
  • Assignment: Saving and loading models
• [Week 5]
  • Capstone Project: Image classifier for the SVHN dataset

Course - 2

Customizing your models with TensorFlow 2 - Coursera

Table of Contents

• [Week 1]
  • Lesson Topic: The Keras functional API, Variables and Tensors, Accessing model layers, Layer nodes, Freezing layers, Device placement (CPU, GPU)
  • Assignment: Transfer learning
• [Week 2]
  • Lesson Topic: Data Pipeline, Keras datasets, Dataset generators, Image data augmentation, Data generators for time series, Introducing the tf.data module, Creating Dataset objects from other data sources, Training with Datasets, TensorFlow Datasets
  • Assignment: Data pipeline with Keras and tf.data
• [Week 3]
  • Lesson Topic: Sequence Modeling, Preprocessing sequence data, Tokenizing text data, Embeddings, Recurrent neural networks, Stacked and bidirectional RNNs, Stateful RNNs
  • Assignment: Language model for the Shakespeare dataset
• [Week 4]
  • Lesson Topic: Model subclassing, Custom layers, Allowing flexible inputs for custom layers, Automatic differentiation, Custom training loops, Tracking metrics in custom training loops, Optimising performance with tf.function
  • Assignment: ResNet Residual network
• [Week 5]
  • Capstone Project: Neural translation model

Course - 3

Probabilistic Deep Learning with TensorFlow 2 - Coursera

Table of Contents

• [Week 1]
  • Lesson Topic: TensorFlow Distributions, Univariate distributions, Multivariate distributions, The Independent distribution, Broadcasting rules, Sampling and log probs, Trainable distributions
  • Assignment: Naive Bayes and logistic regression
• [Week 2]
  • Lesson Topic: Probabilistic layers and Bayesian neural networks, Maximum likelihood estimation, The DistributionLambda layer, Probabilistic layers, Bayes by backprop, The DenseVariational layer, Reparameterization layers
  • Assignment: Bayesian convolutional neural network
• [Week 3]
  • Lesson Topic: Bijectors and normalizing flows, Scale bijectors and LinearOperator, The TransformedDistribution class, Subclassing bijectors, Normalising flows
  • Assignment: RealNVP
• [Week 4]
  • Lesson Topic: Variational autoencoders, Encoders and decoders, Kullback-Leibler divergence, Maximising the ELBO, KL divergence layers
  • Assignment: Variational autoencoder for Celeb-A
• [Week 5]
  • Capstone Project: Probabilistic generative models