Powering AWS purpose-built machine learning chips. Blazing fast and cost effective, natively integrated into PyTorch and TensorFlow and integrated with your favorite AWS services
Role in this project:
ML Engineer Contributions:39 commits, 6 PRs, 8 pushes in 10 months
Contributions summary:Nik primarily contributed to benchmarking and adapting a PyTorch BERT model for Inferentia hardware within the AWS Neuron SDK. Their work involved setting up and testing the model on the MRPC dataset, including code for dataset creation, results reporting, and parallel inference execution. The user also made modifications to the existing BERT tutorial notebooks and scripts, and corrected formatting and paths to improve usability.
awsaws-servicesmachine-learningpytorchtensorflow
Tensors and Dynamic neural networks in Python with strong GPU acceleration
Contributions:8 pushes, 1 branch in 1 year 4 months