Setting the Roadmap

DPS911 AKA Open Source 2 has now begun, and with that comes lots of excitement, ideas, but most importantly planning. With the Co-op applications now beginning, and deadlines looming, it was very clear which field I wanted to focus on; Machine Learning.

There are thousands upon thousands of open source projects on GitHub, however I believe it is important to find a project with many issues available, as well as active, and this points me in the direction of making contributions to a machine learning library, rather than a project that implements machine learning. On top of this, I will be working in Python as that is one of the leading machine learning languages, as well as a language I am very comfortable with.

The Candidates

TensorFlow

  • 1,624 open issues
  • 1,782 contributors
  • 118,000 stars

TensorFlow is an open source software library for numerical computation using data flow graphs.

This is a ML library that I have used in the past working on an applied research project for Seneca, and is an incredibly powerful tool that I would love to be able to contribute to.

Scikit-learn

  • 1,184 open issues
  • 1,229 contributors
  • 32,000 stars

Scikit-learn is a holy grail of ML libraries. This single library implements and combines various ML algorithms to create one universal library for algorithms such as; classification, regression, clustering, dimensionality reduction, model selection, and preprocessing

Keras

  • 2,036 open issues,
  • 765 contributors
  • 37,000 stars

Keras is actually a ML library that I have not only used, but also contributed to! In my previous open source releases, I made a contribution to Keras, although it was relatively small as it was purely documentation.

Next Steps

From here my journey begins in trying to get involved with any or all of these projects to get a feel of the community, as well as where I could make contributions. All of these projects will involve me facing the inaugural learning curve, but my mindset is to stay with one project through all 6 releases, therefore the learning curve will only need to be done once. At this moment I do not have a preference, but will make my selection soon!

 

Until next time…

 

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s