How To Deploy Machine Learning Models Using FastAPI-Deployment Of ML Models As API’s
Вставка
- Опубліковано 8 лют 2025
- github :github.com/kri...
FastAPI is a modern, fast (high-performance), web framework for building APIs with Python 3.6+ based on standard Python type hints.Sebastian is the creator of FastAPI, an open-source Python web framework for building production-ready APIs quickly and easily: with autocompletion everywhere in your editor, automatic type checks, as little code as possible, automatic API documentation, and several other features.
⭐ Kite is a free AI-powered coding assistant that will help you code faster and smarter. The Kite plugin integrates with all the top editors and IDEs to give you smart completions and documentation while you’re typing. I've been using Kite for a few months and I love it! www.kite.com/g...
All Playlist In My channel
FASTAPI playlist: • FastAPI Deployment Tut...
Interview Playlist: • Machine Learning Inter...
Complete DL Playlist: • Complete Road Map To P...
Julia Playlist: • Tutorial 1- Introducti...
Complete ML Playlist : • Complete Machine Learn...
Complete NLP Playlist: • Natural Language Proce...
Docker End To End Implementation: • Docker End to End Impl...
Live stream Playlist: • Pytorch
Machine Learning Pipelines: • Docker End to End Impl...
Pytorch Playlist: • Pytorch
Feature Engineering : • Feature Engineering
Live Projects : • Live Projects
Kaggle competition : • Kaggle Competitions
Mongodb with Python : • MongoDb with Python
MySQL With Python : • MYSQL Database With Py...
Deployment Architectures: • Deployment Architectur...
Amazon sagemaker : • Amazon SageMaker
Please donate if you want to support the channel through GPay UPID,
Gpay: krishnaik06@okicici
Telegram link: t.me/joinchat/...
Please join as a member in my channel to get additional benefits like materials in Data Science, live streaming for Members and many more
/ @krishnaik06
Please do subscribe my other channel too
/ @krishnaikhindi
Connect with me here:
Twitter: / krishnaik06
Facebook: / krishnaik06
instagram: / krishnaik06
#FASTAPI
I stumbled upon your channel and I think it is one of the best things I have seen recently. Thank you for making things so simple to understand.
Thanks sir for all you do. I hope the next video on this series will be on deploying machine learning model on cloud platform usingfastapi
This tutorial helped me in right time. Thank you krish sir.
Great sir Fast Api is much simple than Flask Frame work
Very good explanation. Detail oriented and covering all necessary details. Great Video.
Excellent tutorial krish. Please continue incremental learning playlist
thank you, I always wanted to learn fastAPI. DREAM COME TRUE. Pls.... make ur nxt video soon, I want to learn how to connect API with front-end
very nice and useful video for beginers
This is so clear. Great example.
In detailed explanation. Thank you sir.
learned a lot from you waiting for your next video
thanks for guidance..i'll apply your approach
done with my TensorFlow deep learning model for custom image classification , created fast API to test it in local, now time to deploy it in AWS
How to do that bro
Thank you for this tutorial
Asynchronous server gateway interface
Nailed It. Can I Get the video link of Deploying Deep Learning model using FastAPI??
You are great sir
Good evening sir , it's not correct time to ask this question but it's important for me
Is SAP is good for career I mean for long term
If yes please give me reply sir
It is important for me
It's not worth it!
Verry nice sir how to depoly cotton disease prediction model using fast api
Thank you for the session.
Question: how to work with pydantic when we need to handle exceptions? For example right now you are taking 4 features, but what if any feature is missing in the input?
Very nice subject
very good session .greet like as boos ...
Getting this error when giving command:uvicorn app:app --reload in anaconda prompt. Using vscode
Error:
classifier=pickle.load(pickle_in)
ModuleNotFoundError: No module named 'sklearn.svm._classes'
Same here
Have you solved it?
@@anaswahid8520 Pickle is causing error. I used joblib instead.
Save and Load the model with *joblib* :
filename = 'saved_joblib_model.sav'
loaded_model = joblib.load(filename)
Sir please continue uploading fastapi videos.
Really thanks!!
good video just one question what to do when you have large json file with same data how to talk with API when you have a lot of client
very very helpful
Does it work the same for Mac os
Thank you
Please how can I go about deploying CSV files with categorical features. When I used the same workflow on my model,I am getting a value error: could not convert "France" to integer. I have done preprocessing with LabelEncoder and OneHotEncoder but still get same error. Please help out!
Sir can you please help us in knowing how to consume this api in other applications??
Hi Krish. Nice video. One query, you created the model in Jupyter note book. How did you take that to FAST API. Do we need to make a pickle file and then use in in FAST API? Thanks
Yup. First save the model as a pickle file then u can use it in FastAPI by importing pickle
Can we deploy two different models at a time?
Can we use file up to predict if it's a bank note through the api
What do you recommend if you also want to incorporate html and css? Python Eel?
krish pls make next video in which we are using html as our front end for showing result instead of swagger api. pls make this vedio ASAP as i can then deploy it on cloud and can add link into my resume
This is highly required. PleASE MAKE A VIDEO
Hi, i can't to check predict from browser without swagger... Can you help me?
Gate name? Please explain
Which ide is this which is of black colour?
Spyder
What if I want to pass image in POST? what will be basmodel function then?
Thank you Krish for all your tutorials. Please, can someone help me. I made machine learning models in its own environment in anaconda and I want to deploy them using flask but I cant get either pickle or joblib to work in anaconda environment. Does anyone know what I'm doing wrong please?
Can someone tell me where is he loading the model , is he loading it on the local machine ?
how can we store the predicted values in database?can somebody please help with thiz?
Krisk i have this error : RuntimeError: uvloop does not support Windows at the moment
How can i fix it ?
Hi sir, I have deployed yolov4 using darknet on my local machine. Now I want to deploy it using FastAPI. I just know what is what in FastAPI and just saw this video. Can you please suggest tips so that I can successfully deploy it?
have you get some info? I deploy yolor on Django, but I think the process runs slowly, I want to probe FAST API
Krish, will Pickle file format works for Keras CNN's aswell??
Very nice content but i think you should speak slowly for most easier understanding
How to deploy this model as html. As u deploy model using flask.
Nice thnk you
Sir, is it okay to learn fast api even if I don't know django or flask , i only learned ML till now, please suggest,
Thank you krish sir
Can you make a video on making a fast api for Pytesseract for OCR . It would be helpful
Hey,
How do i use a data member of type datetime in BankNote basemodel class?
I have similar issues, do you have an idea how I can fastapi with type str in bank Chun_Customer model?
I saved my model in h5 format. Now I want load the model using "model=tf.keras.models.load_model(filepath.h5)" in FastAPI. But I am getting error while running API using conda virtual env. What should I do? @Krish
What if the pickl file is large?
Sir can you please make Probability for Machine Learning Playlist. I watch your Statistics for Machine Learning and it is very helpful to me but I can't find the playlist for Probability for Machine Learning. So Please Sir can you make playlist for Probability for Machine Learning.
Hi Sir you did great explain but you said at end of video you will continue to upload deploy deep learning model and and return HTML file continuation not posted till please make this playlist remaining video to finish
Can i create these api' sin vs code?
I don't know what you mean by deploy, but I dont think deploy means running it on localhost.
somehow the prediction part got an Internal Server Error. not sure if anyone else had the same issue.
same for me, I used other models such as SVC and it works.
When will AI-IOT, Augumanted-Reality,Virtual-Reality iNeuron Community batch eagerly waiting
sir please make a video on model deployment by FastApi on jupyter notebook IDE.
Nice !
Krish sir, please make a video of *"Deploying Streamlit on GCloud"* because Github is not supporting files more than 100MB but my model.h5 file is more than 100MB. Please make a tutorial on it... 😐😐😐😐😐
Use GIT LFS for that.
Cool
Asynchronous Server Gateway Interface
when you uploading deployment of deep learning model using this fastapi
First Comment
Hello,
I have a query, hoping for a reply .
I have aYOLOv3 Model, Deployed using Fast API but the issue I am getting is; when I call the model through API the inferencing detection accuracy decreasing.
for example :
If a image have 5 object of a class.
If I am inferencing without API- I could detect all the Objects.
If I am inferencing using API- I miss 1 or 2 objects
Could you please help with the possible reason and how to overcome.
Thankyou
Rama
Asynchronous service gateway interface...
asynchronous server graphical interface
gateway not graphical
I couldn't understand anything 😕; it was a poorly designed course for beginners learning FastAPI for the first time. 👎
second
Asynchronous server gateway interface