The Security Risks of Pickle in Machine Learning
Pickle has been the de facto format for ML models primarily because of its ease of use, native Python support, and (questionable) portability. But it is also extremely insecure due...
Read more →The Blog
TorchStudio: A Local IDE for PyTorch
Got a chance to try out TorchStudio, a local IDE purpose built for PyTorch. While still in early development, it has an impressive feature list that includes analyzing datasets, building...
Read more →AWS Summit San Francisco 2022: ML Training at Scale with SageMaker
If large-scale models and datasets pique your interest, join Gal Oshri and me at the AWS Summit San Francisco for a deep dive session on machine learning training at scale...
Read more →AWS Machine Learning Week: Starting Your ML Journey
Democratizing machine learning starts with making ML easy to use and easy to learn. Join Américo Carvalho and me at AWS Machine Learning Week for a session on Starting your...
Read more →Debugging Neural Networks with SageMaker Studio Lab's Visual Debugger
Debugging even a basic neural network can get real messy, real fast. Did you know SageMaker Studio Lab is based on JupyterLab 3 with a built-in visual debugger? Set breakpoints,...
Read more →