AI Quality Control (AIQC); Framework for Reproducible and Rapid Deep Learning

Welcome to our third talk of 2021! This month we're collaborating with PyData Boston (https://www.meetup.com/PyData-Boston-Cambridge/) to have Layne Sadler presenting on an open source framework for AI Quality Control (see full abstract below) plus breakout rooms for networking.

Zoom link below

NOTE DATE CHANGE: We've changed the date from March 23 to March 18.

== Agenda ==

7:00pm: Introductions
7:15pm: Breakout Rooms
7:30pm: Talk by Layne Sadler
8:00pm: Q&A
8:15pm: Breakout Rooms
8:30pm: End of Event

== Abstract ==

The purpose of this talk is to demonstrate how researchers can use the AIQC framework (https://github.com/aiqc/aiqc) to easily integrate deep learning into their work. AIQC is an open source Python package that simplifies: data preparation, tuning models in batches, and generating performance metrics/ plots. It provides a high-level API that reduces the amount of code needed to perform best practice machine learning by 90%.

Here are some examples of what it enables:

• Performing dtype-specific, leakage-free, and stratified encoding on cross-folded tabular/ image datasets is as simple as specifying `fold_count=5, feature_encoders={}`.
• Similarly, you can specify a dictionary of hyperparameters to pass into a batch of model training jobs. Performance metrics and plots are automatically generated for all splits/ folds based on the type of analysis conducted. If you want to repeat non-deterministic training runs multiple times just set `repeat_count=3`.
• All experiments are automatically recorded in a local SQLite database.

By lowering the technical and data science know-how barriers to deep learning, we can drive its adoption in scientific fields to accelerate Earth-saving discoveries.

== Speaker Bio ==

An autodidact at heart, Layne began on the business side of technology, but was drawn toward building applications and algorithms alike. This led to founding an API-driven startup (athlete.studio) and spearheading product development at a biotech (Genuity Science). While working with pharma and research institutes on national genomic biobank projects, he observed major gaps that prevented the adoption of deep learning in scientific research. So he built AIQC (https://github.com/aiqc/aiqc) to address those problems.

== Sponsors ==

We thank MassMutual for their continued support of the Meetup. Check out the new MassMutual Data Science website here: https://datascience.massmutual.com/