A real-time inference server
Project description
model server
---
title: REST-inference service
---
classDiagram
note "100 requests per second"
class VenueRating{
"""
Represents the predicted ranking of a venue.
Attributes:
-----------
venue_id : int The ID of the venue being rated.
q80_predicted_rank : float
The predicted ranking of the venue,
as a 80-quantile of predicted rating
for venue across available sessions
"""
venue_id: int
q80_predicted_rank: float
}
class TrainingPipeline{
str pre-trained-model-file: stored with mlflow in gcs bucket
}
class InferenceFeatures{
venue_id: int
conversions_per_impression: float
price_range: int
rating: float
popularity: float
retention_rate: float
session_id_hashed: int
position_in_list: int
is_from_order_again: int
is_recommended: int
}
class FastAPIEndpoint{
def predict_ratings(): Callabe
}
class Model_Instance{
joblib.load(model_artifact_bucket)
str model_artifact_bucket - variable
str rank_column - fixed for the model
str group_column - fixed for the model
}
TrainingPipeline --|> Model_Instance
InferenceFeatures --|> FastAPIEndpoint
Model_Instance --|> FastAPIEndpoint
FastAPIEndpoint --|> VenueRating
Documentation: https://ra312.github.io/model-server Training Source Code: https://github.com/ra312/personalization Source Code: https://github.com/ra312/model-server PyPI: https://pypi.org/project/model-server/
A model server for almost realtime inference
Installation
pip install model-server
Development
- Clone this repository
- Requirements:
- Poetry
- Python 3.8.1+
- Create a virtual environment and install the dependencies
poetry install
- Activate the virtual environment
poetry shell
Testing
pytest
Documentation
The documentation is automatically generated from the content of the docs directory and from the docstrings of the public signatures of the source code. The documentation is updated and published as a Github project page automatically as part each release.
Releasing
Trigger the Draft release workflow (press Run workflow). This will update the changelog & version and create a GitHub release which is in Draft state.
Find the draft release from the GitHub releases and publish it. When a release is published, it'll trigger release workflow which creates PyPI release and deploys updated documentation.
Pre-commit
Pre-commit hooks run all the auto-formatters (e.g. black
, isort
), linters (e.g. mypy
, flake8
), and other quality
checks to make sure the changeset is in good shape before a commit/push happens.
You can install the hooks with (runs for each commit):
pre-commit install
Or if you want them to run only for each push:
pre-commit install -t pre-push
Or if you want e.g. want to run all checks manually for all files:
pre-commit run --all-files
This project was generated using the wolt-python-package-cookiecutter template.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for recommendation_model_server-0.0.1.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9ff1e3e9d48cec03d15a868b154db4446d35576bf197810ed382e67c85a096cb |
|
MD5 | 2d81cc25af9ddb64c2c3011dc710b2be |
|
BLAKE2b-256 | 346da36b1d95a237e5b2015a011182bbf1d1b75f471acc34381e7ea6aa308cca |
Hashes for recommendation_model_server-0.0.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 125adde44a03b346222a14ce12f37835d121146e6a5e977fc1e9348f691c09bd |
|
MD5 | 1f1b977c19c0ed418b53d0f41c3f9e64 |
|
BLAKE2b-256 | 39d7702e85548cc227a1cfce5da9e53b7939b70df7f5db76530938701590f2ad |