AI Deploy - Tutorial - Deploy an app for sentiment analysis with Hugging Face and Flask

Last updated 24th February, 2023.

Objective

The purpose of this tutorial is to show you how to deploy a web service for sentiment analysis on text using Hugging Face pretrained models.
In order to do this, you will use Flask, an open-source micro framework for web development in Python. You will also learn how to build and use a custom Docker image for a Flask application.

Overview of the app:

For more information about Hugging Face, please visit https://huggingface.co/.

Requirements

• Access to the OVHcloud Control Panel;
• An AI Deploy project created inside a Public Cloud project in your OVHcloud account;
• A user for AI Deploy;
• Docker installed on your local computer;
• Some knowledge about building image and Dockerfile;

We also suggest you do some tests to find out which Hugging Face model is right for your use case. Find examples on our GitHub repository.

Instructions

First, the tree structure of your folder should be as follows:

Find more information about the Flask application here to get ready to use it.

Write the Flask application

Create a Python file named app.py.

Inside that file, import your required modules:

from flask import Flask, jsonify, render_template, request, make_response
import transformers

Create Flask app:

app = Flask(__name__)

Load Hugging Face models:

# create a python dictionary for your models d = {<key>: <value>, <key>: <value>, ..., <key>: <value>}
dictOfModels = {"BERT" : transformers.pipeline('sentiment-analysis', model="nlptown/bert-base-multilingual-uncased-sentiment")}
# create a list of keys to use them in the select part of the html code
listOfKeys = []
for key in dictOfModels :
        listOfKeys.append(key)

Write the inference function:

def get_prediction(message,model):
    # inference
    results = model(message)  
    return results

Define the GET method:

@app.route('/', methods=['GET'])
def get():
    # in the select we will have each key of the list in option
    return render_template("home.html", len = len(listOfKeys), listOfKeys = listOfKeys)

Define the POST method:

@app.route('/', methods=['POST'])
def predict():
    message = request.form['message']
    # choice of the model
    results = get_prediction(message, dictOfModels[request.form.get("model_choice")])
    print(f'User selected model : {request.form.get("model_choice")}')
    my_prediction = f'The feeling of this text is {results[0]["label"]} with probability of {results[0]["score"]*100}%.'
    return render_template('result.html', text = f'{message}', prediction = my_prediction)

Start your app:

if __name__ == '__main__':
    # starting app
    app.run(debug=True,host='0.0.0.0')

Write the requirements.txt file for the application

The requirements.txt file will allow us to write all the modules needed to make our application work. This file will be useful when writing the Dockerfile.

Flask==2.1.0

transformers==4.4.2

torch==1.6.0

Here we will mainly discuss how to write the app.py code, the requirements.txt file and the Dockerfile. If you want to see the whole code, please refer to the GitHub repository.

Write the Dockerfile for the application

Your Dockerfile should start with the FROM instruction indicating the parent image to use. In our case we choose to start from a Python image:

FROM python:3.8

Create the home directory and add your files to it:

WORKDIR /workspace
ADD . /workspace

Install the requirements.txt file which contains your needed Python modules using a pip install ... command:

RUN pip install -r requirements.txt

Define your default launching command to start the application:

CMD [ "python" , "/workspace/app.py" ]

Give correct access rights to ovhcloud user (42420:42420):

RUN chown -R 42420:42420 /workspace
ENV HOME=/workspace

Build the Docker image from the Dockerfile

Launch the following command from the Dockerfile directory to build your application image:

docker build . -t sentiment_analysis_app:latest

Test it locally (optional)

Launch the following Docker command to launch your application locally on your computer:

docker run --rm -it -p 5000:5000 --user=42420:42420 sentiment_analysis_app:latest

Once started, your application should be available on http://localhost:5000.

Push the image into the shared registry

Find the adress of your shared registry by launching this command:

ovhai registry list

Login on the shared registry with your usual OpenStack credentials:

docker login -u <user> -p <password> <shared-registry-address>

Push the compiled image into the shared registry:

docker tag sentiment_analysis_app:latest <shared-registry-address>/sentiment_analysis_app:latest
docker push <shared-registry-address>/sentiment_analysis_app:latest

Launch the AI Deploy app

The following command starts a new app running your Flask application:

ovhai app run --default-http-port 5000 --cpu 4 <shared-registry-address>/sentiment_analysis_app:latest

Go further

• You can also imagine deploying an Object Detection model with Flask in this tutorial.
• Discover an other tool to deploy easily AI models: Gradio. Refer to this documentation.

Feedback

Please send us your questions, feedback and suggestions to improve the service:

• On the OVHcloud Discord server

Inne przewodniki, które mogą Cię zainteresować...