Skip to main content

A platform for managing and serving Machine Learning types.

Project description

Note: HuoguoML is stil under development and can have some unknown issues.

🍲 HuoguoML

When dealing with Machine Learning applications, there is a high management and coordination effort for data scientists, as they have to collaborately analyze, evaluate and update many different models with different metadata on a regular basis. HuoguoML aims to simplify the management process by providing a platform for managing and serving machine learning models. It enables:

  • Individual Data Scientists to track experiments locally on their machine and output models to ML engineers, who then deploy them using HuoguoML's deployment tools.
  • Data Science Teams to set up a HuoguoML tracking server to log and compare the results of multiple data scientists working on the same or a different problem. Then, by setting up a convention for naming their parameters and metrics, they can try different algorithms to solve the same problem and then run the same algorithms again on new data to compare models in the future. In addition, anyone can download and run a different model.
  • ML Engineers to deploy models from different ML libraries in the same way by executing a simple command. Each service is centrally maintained and supports OTA updates. On its own, a HuoguoML Service is based on FastAPI and can be extended with standard FastAPI tools. It supports all use cases, be it storing predictions with middleware or adding new endpoints e.g. for Prometheus.

Installation

HuoguoML can be installed via PyPI. Install the stable version of HuoguoML via PyPI:

pip install huoguoml

or get the development version, which is updated with every commit on the main branch:

pip install huoguoml-dev

Examples

Just starting out? Try out our examples which work out of the box:

Example Description
Tensorflow MNIST Building a MNIST classifier with Tensorflow and HuoguoML

Documentation

Apart from learning from the examples, we highly recommended you go through our documentation, as it gives you a more detailed guide to HuoguoML

Our docs are built on every push to the main or docs branch.

Contributing

We encourage you to contribute to HuoguoML! Please check out the Contributing guide for guidelines about how to proceed.

License

Apache License Version 2.0, see LICENSE

Project details


Download files

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

Source Distribution

huoguoml-dev-1624454207.tar.gz (18.9 kB view details)

Uploaded Source

Built Distribution

huoguoml_dev-1624454207-py3-none-any.whl (2.3 MB view details)

Uploaded Python 3

File details

Details for the file huoguoml-dev-1624454207.tar.gz.

File metadata

  • Download URL: huoguoml-dev-1624454207.tar.gz
  • Upload date:
  • Size: 18.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.1 importlib_metadata/4.5.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.7.10

File hashes

Hashes for huoguoml-dev-1624454207.tar.gz
Algorithm Hash digest
SHA256 c834582a5800cf093b679620505a3883f2cb018f9dba18af7b3c3322a13c1463
MD5 cb4d992d5572f64878ffa99b053d7e04
BLAKE2b-256 7a602e412b735b8e18a2676ac73eb4e4584f0afddb2c2ae6be8c041ec885d75d

See more details on using hashes here.

File details

Details for the file huoguoml_dev-1624454207-py3-none-any.whl.

File metadata

  • Download URL: huoguoml_dev-1624454207-py3-none-any.whl
  • Upload date:
  • Size: 2.3 MB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/3.4.1 importlib_metadata/4.5.0 pkginfo/1.7.0 requests/2.25.1 requests-toolbelt/0.9.1 tqdm/4.61.1 CPython/3.7.10

File hashes

Hashes for huoguoml_dev-1624454207-py3-none-any.whl
Algorithm Hash digest
SHA256 2b661e1ce4af3479a6b1186e0f410b99a2af78c29d81172a092c9c0926ebe6dd
MD5 ee274f8ed3cb51628855977d5fb5e806
BLAKE2b-256 76bacd12873c7f5bf6a45d61e688ab2da9f2ead6c0cd8ece126abb1754d139e0

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page