Easy model distribution for inference.

Project description

modelhouse

Why?

The purpose of modelhouse is to make it easy to distribute models to workers on cloud.

With modelhouse, you can store and load your models from different storage backends (Google Cloud Storage, AWS S3, local). modelhouse also provides model caching, which can significantly reduce model loading time while keeping memory utilization low.

Installation

pip install modelhouse

Documentation

TODO

Getting started

TODO

Project details

Release history Release notifications | RSS feed

0.1.4

Jan 19, 2023

0.1.2

Dec 4, 2021

0.1.1

Oct 26, 2021

0.1.0

Oct 26, 2021

0.0.6

Apr 20, 2021

0.0.5

Jun 9, 2020

0.0.3

May 16, 2020

0.0.2

May 16, 2020

This version

0.0.0.1

May 4, 2020

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

modelhouse-0.0.0.1.tar.gz (1.2 kB view hashes)

Uploaded May 4, 2020 Source

Built Distribution

modelhouse-0.0.0.1-py3-none-any.whl (5.5 kB view hashes)

Uploaded May 4, 2020 Python 3

Hashes for modelhouse-0.0.0.1.tar.gz

Hashes for modelhouse-0.0.0.1.tar.gz
Algorithm	Hash digest
SHA256	`8629613b6ecf65b8a49cec7b29bc0a1e50fc0ab0ce1bdc2a0e716e806756e119`
MD5	`acfc263e58f50947d8a519dec8babb33`
BLAKE2b-256	`4841d1dd165ba7ecc345f09fa2b5bbf538741c76155f464307283d72faae54f2`

Hashes for modelhouse-0.0.0.1-py3-none-any.whl

Hashes for modelhouse-0.0.0.1-py3-none-any.whl
Algorithm	Hash digest
SHA256	`7e4999ff2a0ceed3040433115902767c2b90ae0d03dfd9c5a26f005a8d8ff3cf`
MD5	`deab30bea5bc7f9c8bc321c0e5003118`
BLAKE2b-256	`e50e0120c916256ebeb8cf4732e1e9f004440f52ff660c72801731d55e8b66c2`