0
I Use This!
Inactive
Analyzed about 14 hours ago. based on code collected 1 day ago.

Project Summary

Amazon DSSTNE: Deep Scalable Sparse Tensor Network Engine
DSSTNE (pronounced "Destiny") is a library for training and deploying deep neural networks using GPUs. It is build to solve deep learning problems at Amazon's scale. It is built for production deployment of real-world deep learning applications, emphasizing speed and scale over experimental flexibility.
Multi-GPU Scale: Training and prediction both scale out to use multiple GPUs, spreading out computation and storage in a model-parallel fashion for each layer.
Large Layers: Model-parallel scaling enables larger networks than are possible with a single GPU.
Sparse Data: DSSTNE is optimized for fast performance on sparse datasets. Custom GPU kernels perform sparse computation on the GPU, without filling in lots of zeroes.

Tags

deep_learning deployment GPU machine_learning multi-gpu neural_networks production scalable sparse

In a Nutshell, Deep Scalable Sparse Tensor Network E...

Apache License 2.0
Permitted

Commercial Use

Modify

Distribute

Place Warranty

Sub-License

Private Use

Use Patent Claims

Forbidden

Hold Liable

Use Trademarks

Required

Include Copyright

State Changes

Include License

Include Notice

These details are provided for information only. No information here is legal advice and should not be used as such.

This Project has No vulnerabilities Reported Against it

Did You Know...

  • ...
    use of OSS increased in 65% of companies in 2016
  • ...
    check out hot projects on the Open Hub
  • ...
    in 2016, 47% of companies did not have formal process in place to track OS code
  • ...
    data presented on the Open Hub is available through our API

Languages

C++
49%
CUDA
37%
Java
8%
7 Other
6%

30 Day Summary

Feb 26 2024 — Mar 27 2024

12 Month Summary

Mar 27 2023 — Mar 27 2024

Ratings

Be the first to rate this project
Click to add your rating
  
Review this Project!