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.
← Back to Blog
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 for training.
Read article →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
Supercharging Hyperparameter Tuning with Dask
The distributed computing framework Dask is great for hyperparameter tuning, since you can train different parameter sets concurrently.
Practical Issues Setting up Kubernetes for Data Science on AWS
Data science has unique workflows that don't always match those of software engineers and require special setup for Kubernetes.
Setting Up Your Data Science & Machine Learning Capability in Python
Python is a great language to base your DS/ML framework on, and allows you to avoid being locked into one vendor specific framework.