ML Bootcamp 2019

You can find the resources, notes and links that will be used at the Machine Learning Bootcamp here.

Day 1

• Intro to ML
• Problem framing
• Data preparation and Feature Engineering
• Linear and Logistic regression with scikit learn (theory + hands on)
  • Linear - https://colab.research.google.com/drive/1ekd7raM6lBgRYzrQTntXD1gegHlubXYK
  • Logistic - https://colab.research.google.com/drive/1Qva6AwESqgjHkviIVwjUgy-NGRXoa-R0
• Intro to Tensorflow (sessions, graphs and basic principles + hands on)
  • TF Basic - https://colab.research.google.com/drive/1-pIkqipJKiYU01cH3U2ndrCnnaaL_0hR
• Loss function / Gradient Descent / Backpropagation
• First Neural Net with Tensorflow and Intro to Keras(training, testing with different hyper params, loss functions etc)

Lunch Break

• Generalizing (overfitting, underfitting, regularization etc.)
• Feature crossing
• TF Estimators
• Practical exercise : New York taxi fare prediction

Day 2

• Recap of day 1
• New York taxi fare prediction solution
• Embeddings
• How to use GCP for ML(train New york taxi fare model on ML Engine)

Lunch Break

• Convolutional neural networks (practical: an image classification)
  • https://colab.research.google.com/drive/1Kof7J312znH1NAMrjMhpbh8fZ0AXxSS2
• Artistic style transfer
• Basic intro to RNN/LSTMs and their use cases
• Convert the image classification model (built above) and deploy on Firebase
• ML Qwiklabs