Skip to main content

Arrow Flight simple API service

Project description

Aflight: simple Arrow Flight API builder/runner

Loosely inspired by Flask

TL;DR

Server:

Create app.py

import aflight

app = aflight.Aflight()

@app.command("predict")
def predict(batch):
    result = do_something_with(batch)
    return [result]

To Run:

aflight

Client:

from aflight.client import predict

data = some_pandas_data()
print(predict(data))

Why?

Current kserve, torch/serve etc. APIs are too reliant on HTTP request/response paradigm

Using Arrow as a data format and Arrow Flight as a service API produces a much better interface:

  • standard
  • has schema
  • fast
  • columnar
  • batch-able

Project details


Download files

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

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

aflight-0.2.4-py3-none-any.whl (7.4 kB view details)

Uploaded Python 3

File details

Details for the file aflight-0.2.4-py3-none-any.whl.

File metadata

  • Download URL: aflight-0.2.4-py3-none-any.whl
  • Upload date:
  • Size: 7.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/4.0.2 CPython/3.9.16

File hashes

Hashes for aflight-0.2.4-py3-none-any.whl
Algorithm Hash digest
SHA256 436217ddff6390afe37f691f09cef54cdeac1c7f992e38f93a9fb58307cebb32
MD5 c1498599e063b13be99dcae590026fcb
BLAKE2b-256 fb48b0697b34868188293934aff0ddf2a80ef29460679f3fa4794fa9338d70f0

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