Notes and code for the Machine Learning Engineer Nanodegree Program (MLND) by Udacity. The goal of the program is to teach key skills in the area of machine learning. The following excerpt is taken from the program syllabus:
A graduate of this program will be able to:
- Test Python code and build a Python package of their own.
- Build predictive models using a variety of unsupervised and supervised machine learning techniques.
- Understand cloud deployment terminology and best practices.
- Use Amazon SageMaker to deploy machine learning models to production environments, such as a
web application or piece of hardware.
- A/B test two different deployed models and evaluate their performance.
- Utilize an API to deploy a model to a website such that it responds to user input, dynamically.
- Update a deployed model, in response to changes in the underlying data source
Project Overview:
- Build a Python Package: Write a Python package on your own using software engineering best
practices for writing production level code.
- Deploy a Sentiment Analysis Model: Using SageMaker, deploy your own PyTorch sentiment
analysis model, which is trained to recognize the sentiment of movie reviews (positive or negative).
- Plagiarism Detector: Engineer features that can help identify cases of plagiarism in text and deploy
a trained plagiarism detection model using Amazon SageMaker.
- Capstone Project & Proposal: Complete a final project—choosing from a few, provided options or a
project of your own design—that involves data exploration and machine learning.