Introduction
The main goal of the TensorShare package is to provide easy to use and clear interfaces for sharing tensors between different applications connected to a network.
The package is designed to be used in a distributed environment where multiple applications run on different
machines and need to share tensors. The tensors could be in a different backend (e.g. numpy and torch)
and the package will help you to share them simply and transparently.
Tip
The project leverages the best open-source tools and integrates them into a single package to provide you a fast and reliable way to share your tensors.
Main Components
There are 3 main components in the TensorShare package:
- TensorShare - the main schema for sharing tensors across the network and using multiple backends (See Backends)
- TensorShareServer - a server configuration schema defining how tensors could be shared over the network.
- TensorShareClient - the HTTP client for sending tensors to the server.
Easy FastAPI Integration
We also provide tools to integrate TensorShare into your FastAPI application in a matter of seconds. See FastAPI Integration for more details.
Issues and Feature Requests
If you have any issues or feature requests, please feel free to open an Issue on the GitHub repository.
Created: 2023-08-20