aiXplain SDK adds AI functions to software.
Project description
aiXtend
aiXtend enables python programmers to add AI functions to their software.
An aiXplain pipeline is a directed graph (DAG) of AI functions built using aiXplain's designer UI. An AI function is a data processing step that relies on a machine learning model to execute. An example of an AI function is speech recognition or machine translation. Pipelines help you process your data by calling a series of functions as defined in the DAG, abstracting the orchestration, and providing a simple python function call.
aiXplain has a collection of AI models for each AI function. You can explore the collection of our AI models by using the discover feature of our platform's website. You can run a model using simple python code as shown in this example.
aiXplain Pipeline Designer DAG
The image below shows a sample aiXplain pipeline built for subtitling video files. The description of the pipeline can be found in the documentation.
Installation
pip install aixtend
User Guide
In order to use aiXplain, you need to create an account in aiXplain platform. Follow the code samples listed below to get started.
Code Samples and Demos
aiXtend provides python APIs to call AI workflows you can build with aiXplain designer and to call models you subscribed with aiXplain discover.
Pipeline Snippet
Synchronous
from aixtend.factories.pipeline_factory import PipelineFactory
api_key = <API_KEY>
pipeline = PipelineFactory.initialize(api_key=api_key)
path = <DATA_URL>
response = pipeline.run(data=path)
Asynchronous
import time
from aixtend.factories.pipeline_factory import PipelineFactory
api_key = <API_KEY>
pipeline = PipelineFactory.initialize(api_key=api_key)
path = <DATA_URL>
response = pipeline.run_async(data=path)
if response['status'] != 'FAILED':
poll_url = response['url']
completed = False
while not completed:
response = pipeline.poll(poll_url)
completed = response['completed']
time.sleep(3)
Model Snippet
Synchronous
from aixtend.factories.model_factory import ModelFactory
api_key = <API_KEY>
model = ModelFactory.initialize(api_key=api_key)
path = <DATA_URL>
response = model.run(data=path)
Asynchronous
import time
from aixtend.factories.model_factory import ModelFactory
api_key = MODEL_API_KEY
model = ModelFactory.initialize(api_key=api_key)
path = <DATA_URL>
response = model.run_async(data=path)
if response['status'] != 'FAILED':
poll_url = response['url']
completed = False
while not completed:
response = model.poll(poll_url)
completed = response['completed']
time.sleep(3)
API_KEY can be obtained by creating a pipeline in pipeline designer through the aiXplain platform UI.
For DATA_URL generate a http(s) link to your image or video file to process, though text input can be directly supplied to data parameter in the run function without needing a URL.
Information on how to generate the API_KEY can be found in the subtitle generation pipeline sample video.
Subtitle Generation
This demo creates a .srt file for the supplied video using aixplain-pipelines. Follow the instructions in the documentation.
Developer Guide
Follow the developer guide documentation.
Support
Raise issues for support in this repository.
Pull requests are welcome!
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 aixtend-0.0.5.macosx-10.9-x86_64.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 71341d056dceec16d071dc77085805c9946fdcc815c94f3b9691dea8e2b98548 |
|
MD5 | 7d75389b026e6205ff05ddbe8ec959d0 |
|
BLAKE2b-256 | 104cc5888a415ae9aa2c475191e9bcd073278523db2bb55eaa55dc607a7ffa07 |