Meeshkan: Machine Learning the GitHub API

Name: Meeshkan: Machine Learning the GitHub API
Rating: 4.0 (95 reviews)

Learn how to plan, deploy and run a Machine Learning problem on AWS and Meeshkan

Created byMike Solomon

Last updated 1/2018

English

What you'll learn

Collect and analyze a large dataset using Meeshkan.

Course content

2 sections • 15 lectures • 1h 32m total length

Welcome!1:24
Meeshkan: Machine Learning the GitHub API12:08
Welcome to Meeshkan: Machine Learning the GitHub API!
In this video, we are going to do the whole course from beginning to end in miniature.
Every subject of this video: building a data collecting algorithm, serving it via a webhook, deploying it to AWS, designing a model, running the model on Meeshkan and visualizing the results, will be the subject of separate videos where we will explore each topic in depth.
And don't worry if you feel that it goes by fast, it's normal! Eventually, you'll be this fast with this stuff too. The most important thing is to realize that working with AI can be fast, fun and easy if you have the right tools.
So for this video, sit back, relax and enjoy the overview! You'll get some hands-on experience starting from the second lesson.

Setting Up Your Data Collection Environment0:37
Collecting Data21:52
Collect data from the GitHub public API using the saga pattern to create a predictable, scalable, and testable data collection workflow; deployable with the open source data server.
Downloading the Webhook Source Code0:27
Serving Data via a Webhook16:07
Learn how to build a webhook to serve GitHub API data for machine learning, implement data partitions, and use author and committer windowing with normalized features.
Getting Familiar With YAML0:28
Deploying Infrastructure11:03
Deploy infrastructure for data collection and webhook using infrastructure as code. Address GitHub API rate limits and database bottlenecks with scalable AWS resources and cost considerations.
Machine Learning and Keras0:43
Designing Your Models12:31
Design and train simple dense neural network models with Keras on Meeshkan, leveraging backends like Torch or TensorFlow, activation functions, and gradient descent to learn from GitHub API data.
Signing Up for Meeshkan0:13
Running Your Models on Meeshkan2:43
Launch machine learning on meeshkan using data from the GitHub public API, queue and deploy models via the dashboard, and tune hyperparameters like epics and mini batches.
Downloading the Data Visualizer0:51
Analyzing Your Models9:00
Explore how to download and visualize meeshkan results, evaluate binary predictions and actual star counts from GitHub data, and analyze model performance using the data visualizer.
You Are the Future of Machine Learning: Go For It!2:05
Design, deploy, and analyze models on the GitHub API within Meeshkan’s sandbox to gain intuition about machine learning. Use ML to ask better questions and explore ideas.

Requirements

You should have basic familiarity with Machine Learning. JavaScript knowledge is a plus!

Description

In this course, Meeshkan C.E.O. Mike Solomon will teach you how to do Machine Learning on Meeshkan.

Meeshkan is an easy and inexpensive platform where people can explore ideas in AI, Machine Learning and Deep Learning.

This course starts with a simple AI question: can a machine predict if a GitHub project will be successful by analyzing only the first few commits of that project?

The first section of the course will run the Machine Learning project on Meeshkan. You'll see how quick and easy it is to do Machine Learning on Meeshkan.

The second section of the course will delve into each step of the process in detail, covering data collection, data egress, infrastructure deployment, model design, model executing and result analysis.

By the end of the course, you will be able to adapt the course materials to design, run, and explore your own Machine Learning models using public APIs and the Meeshkan Machine Learning service.

Who this course is for:

Anyone who wants to learn about Machine Learning.

Meeshkan: Machine Learning the GitHub API

What you'll learn

Explore related topics

Course content

Let's (Machine) Learn!2 lectures • 14min

Machine Learning with Meeshkan - Step by Step13 lectures • 1hr 19min

Requirements

Description

Who this course is for: