Notebook Projects

Deep learning projects to understand the neural network concept, implement neural network and improve the performance of our result.

1. Initialization

   In this notebook project we will:

   • Understand that different initialization methods and their impact on our model performance

   • Implement zero initialization and and see it fails to "break symmetry",

   • Recognize that random initialization "breaks symmetry" and yields more efficient models,

   • Understand that we could use both random initialization and scaling to get even better training performance on our model.

2. Regularization

   In this notebook project we will:

   • Understand that different regularization methods that could help our model.

   • Implement dropout and see it work on data.

   • Recognize that a model without regularization gives us a better accuracy on the training set but nor necessarily on the test set.

   • Understand that we could use both dropout and regularization on our model.

3. Optimization

   In this notebook project we will:

   • Understand the intuition between Adam and RMS prop

   • Recognize the importance of mini-batch gradient descent

   • Learn the effects of momentum on the overall performance of our model

4. Gradient Checking

   In this notebook project we will:

   • Implement gradient checking from scratch.

   • Understand how to use the difference formula to check our backpropagation implementation.

   • Recognize that our backpropagation algorithm should give us similar results as the ones we got by computing the difference formula.

   • Learn how to identify which parameter's gradient was computed incorrectly.

5. Tensorflow

   • In this notebook we will learn all the basics of Tensorflow.
   • We will implement useful functions and draw the parallel with what we did using Numpy.
   • We will understand what Tensors and operations are, as well as how to execute them in a computation graph.