Home

Quickstart

torchlambda allows you to take your PyTorch neural networks to production on AWS Lambda using a few (4 at most) commands.

To jump straight into some model deployment please see the following tutorial: ResNet18 deployment on AWS Lambda. This is the first recommended step for newcomers.

Secondly you may want to check whether your model will work on AWS Lambda. It is a good idea to verify it locally, for that check Testing your deployment locally.

For reference about specific functionalities and use cases see below.

Table Of Contents

• Installation

• Tutorials

  • ResNet18 deployment on AWS Lambda
  • base64 image encoding
  • Testing your deployment locally

• Commands

  • settings
  • template
  • build
  • layer

• YAML settings file reference

• C++ code

Related projects

Other projects from the same author supporting/extending PyTorch capabilities:

• torchlayers - Shape inference (like in Keras) for PyTorch + new SoTA layers!
• torchdata - Extend PyTorch's Dataset with map, cache, filter etc. (like tensorfow.data)
• torchfunc - PyTorch functions & utilities to make your deep learning life easier