Lazy Evaluation with Dask
Lazy evaluation is when calculations are only computed when they are needed, and Dask makes great use of the method.
← Back to Blog
Dask and pandas: There’s No Such Thing as Too Much Data
Do you love pandas, but hate when you reach the limits of your memory or compute resources? Dask gives you the chance to use the pandas API with distributed data and computing.
Read article →Snowflake & Saturn Cloud Launch 100x Faster Machine Learning Platform
Snowflake and Saturn Cloud are partnering to help data science teams get 100x faster results.
Combining Dask and PyTorch for Better, Faster Transfer Learning
PyTorch has the ability to train models across multiple machines, and thanks to the framework Dask you can easily create a GPU cluster …
Handy Dandy Guide to Working With Timestamps in pandas
Using times in pandas can sometimes be tricky--this blog post covers the most common problems.
Computer Vision at Scale With Dask and PyTorch
This tutorial walks through how to use PyTorch and Dask to train an image recognition model across a GPU cluster.
Saturn Cloud Hosted Has Launched: GPU Data Science for Everyone!
Our new Saturn Cloud Hosted platform lets data scientists get going with cloud compute in seconds.
Cross-Entropy Loss Function
When working on a Machine Learning or a Deep Learning Problem, loss/cost functions are used to optimize the model during training. The …
The Future of Computer Vision with AI Pioneer Senseye
Senseye uses Saturn Cloud to train machine learning models on GPUs at a massive scale.
Random Forest on GPUs: 2000x Faster than Apache Spark
This blog post compares using RAPIDS and Dask vs Apache Spark for model training