Spell Blog
MLOps
A developer-friendly guide to model quantization with PyTorch
December 2nd, 2020
Quantized CPU model are up to 75% smaller, with 50% faster inference times
Community
Simple Contrastive Learning Representation using the STL10 dataset in PyTorch
November 30th, 2020
In this blog post, we will cover the Simple Contrastive Learning Representation Framework developed by Google, and implement SimCLR from scratch using PyTorch and PyTorch Lightning.
Scaling ML
Key Concepts for Deploying Machine Learning Models to Mobile
November 18th, 2020
In this guest article from Fritz AI, we'll be outlining the key considerations in building a mobile-focused ML pipeline from end-to-end.
MLOps
Getting oriented in the RAPIDS distributed ML ecosystem, part 2: training and scoring
November 13th, 2020
RAPIDS and RAPIDS accelerates workflows using classical machine learning techniques by a factor of 20x or more.
Product Updates
Accessing AWS services from within Spell using custom instance profiles
November 9th, 2020
A guide to integrating Spell with your existing AWS services
Ready to Get Started?
Create an account in minutes or connect with our team to learn how Spell can accelerate your business.