Petastorm library enables single machine or distributed training and evaluation of deep learning models from datasets in Apache Parquet format. It supports ML frameworks such as Tensorflow, Pytorch, and PySpark and can be used from pure Python code.

Pytorch implementation of GEE: A Gradient-based Explainable Variational Autoencoder for Network Anomaly Detection

This repo contain the code related to the Medium post: https://medium.com/p/168b47e54d54

MNIST with Petastorm on Sagemaker using Tensorflow Estimators

deep learning at scale

Petastorm library enables single machine or distributed training and evaluation of deep learning models from datasets in Apache Parquet format. It supports ML frameworks such as Tensorflow, Pytorch, and PySpark and can be used from pure Python code.

A demo for petastorm and pyspark

Minimal reproducible example to debug Petastorm data loader behavior when interacting with Pytorch lightning

No description available

No description available

manually fork of uber petastorm repo

No description available

Use of PetaStorm

No description available

scripts for figuring out and testing petastorm

No description available

Code to create and read flash match training data using the Petastorm library

No description available

No description available

Databricks DL with PyTorch Lightning and Petastorm

No description available

CheXpert project using CNN with Petastorm ETL management.

Take AWS tensorflow container for SageMaker and add Petastorm

An example implementation of distributed training of transformer models on Databricks using Horovod and Petastorm.

Useful elements and building blocks for scalable Deep Learning applications on Databricks.