Deploy ML Models from Colab with FastAPI & ColabCode - Free ML as a Service

preview_player
Показать описание
In this Python Tutorial, We'll learn how to deploy ML Models from Google Colab with Sebastián Ramírez's {FastAPI} and Abhishek Thakur's {colabcode} Python Libraries.

How to create a simple API in Python with FastAPI
How to save an ML model with Pickle and use the Pickle for Inference
How. to serve/deploy the ML Model Inference as an API with FastAPI
How to use ColabCode to run FastAPI app / API and tunnel it using ngrok URL

Jupyter Notebook -

Packages -
Рекомендации по теме
Комментарии
Автор

help me, how can I give the function get_predictions two parameters of type Upload, of different format, one for the video and the other for the audio?

hafsa
Автор

hello when i use different model the body response is error

michaelianleguira
Автор

Hi, colab can't detect the models.py file that was created. I have also tried adding it to a new folder and pointing the colab in that folder but it is not working. Any idea why? ):

dorianmon
Автор

I'm getting a warning saying "You may be executing code that is disallowed, and this may restrict your ability to use Colab in the future. Please note the prohibited actions specified in our FAQ." when I try to install colabcode...any idea why?

thottytoo
Автор

why it says 'details not found', I have the same issue when running your code, how to fix it?

keyvannoroozikia
Автор

But Google colab have runtime upto 10 hours so for running app in the server for 24 hours we should have to use EC2 instance.

ravikshdikola
Автор

Can i recreate the exact code in Spyder? Or only works on google colab?

DJSamuelLeal
Автор

Have you passed the generated URL in GC to ngrok? Can you possibly tell me how you did that? Just want to compare notes. Thnx.

Lisaiceland
Автор

have I ever wondered?! have I ever wondered!!!
boy!

greendsnow
Автор

colabcode is now banned by Google colab, so your tutorial is obsolete

bqgezck