Python Deep Learning: Parts 1-2

Location: Remote via Zoom. Link will be sent on the morning of the event.

Date & Time:This workshop is a 2-part series running from 9am-12pm:

• Part 1: Tuesday, June 7
• Part 2: Thursday, June 9

Description

The goal of this workshop is to build intuition for deep learning by building, training, and testing models in Python. Rather than a theory-centered approach, we will evaluate deep learning models through empirical results.

We start with a review of what deep learning is and then unpack what neural networks are and how they work. We then jump straight into Python, using the Keras library to build neural networks. We will explore how different architectures affect performance of predicting handwritten digit images.

Lastly, we explore a specific flavor of neural networks, the convolutional neural network. We review how it’s different from a standard vanilla neural network, and build different architectures to test how well they perform on the classification of animal and vehicle image classification.

Prerequisites: D-Lab's Python Introduction to Machine Learning: Parts 1-2 or equivalent knowledge.

This is an advanced level workshop. Participants should be intermediate Python users and have had some prior exposure to machine learning.

We assume the following background:

• D-Lab's Python Machine Learning Fundamentals (6 hours)
• Or, comparable experience/training, assuming familiarity with:
  • Basic Python syntax
  • Train/validation/test splitting
  • Dataset cleaning
  • Overfitting / underfitting / generalization
  • Hyperparameter customization
  • Basic linear algebra (vector, matrix, etc.)
  • Basic statistics (linear regression)

Workshop materials: https://github.com/dlab-berkeley/Python-Deep-Learning

Feedback: After completing the workshop, please provide us feedback using this form