IYKRA Sentiment Analysis

A simple sentiment analysis module wrapped in Flask.

This repo is used as a resource for model deployment sharing at IYKRA (https://iykra.com/).

Getting Ready

Git

• You will use Git (and GitHub) to access this code base and to submit the assignment later.

• Create a GitHub account (https://github.com/)

• Install Git: https://www.atlassian.com/git/tutorials/install-git.

• Learn a bit about Git: https://opensource.com/article/18/1/step-step-guide-git.

• (Optional) If you want to dive deeper: https://towardsdatascience.com/getting-started-with-git-and-github-6fcd0f2d4ac6.

• Fork this repo, follow the instruction here: https://docs.github.com/en/github/getting-started-with-github/fork-a-repo)

• Clone the forked repo by running the following command inside your terminal:

git clone https://github.com/{Your-GitHub-username}/iykra-sentiment-analysis

• It will create iykra-sentiment-analysis directory in your local machine, please remember the path to this directory.

Python

• We're using Python as our main programming language (and some libraries in it).

• Install Python3.6 or above (https://realpython.com/installing-python/).

• Check it, for example you can run: python3 --version or python --version in your terminal.

• You're okay if the output says it's Python3.6 or above.

IDE

• Install your favorite IDE, mine is Atom (https://atom.io/). Atom is very light (got it?)!

• Some other alternatives are:

1. PyCharm (https://www.jetbrains.com/pycharm/)

2. Microsoft Visual Studio (https://visualstudio.microsoft.com/)

Postman

• We will use Postman to test our deployed ML model.

• Install Postman (https://learning.postman.com/docs/getting-started/installation-and-updates/).

Heroku

• Please create a Heroku account: (https://www.heroku.com/).

• (Optional) Install Heroku CLI (https://devcenter.heroku.com/articles/heroku-command-line).

• (Optional) Read these two amazing resources on deploying Flask to Heroku:

1. https://stackabuse.com/deploying-a-flask-application-to-heroku/

2. https://www.jcchouinard.com/deploy-a-flask-app-on-heroku/

• We will follow closely the above resource in the hands on session.

Setting Up Environment

• Inside the iykra-sentiment-analysis directory, please run the following commands to install the libraries:

pip3 install -r requirements.txt

• If the above doesn't work, try:

pip install -r requirements.txt

• (Optional) If you're familiar with virtualenv, run the above command inside your virtual environment.

• Please refer here: https://www.petanikode.com/python-virtualenv/.

• Run the two commands below in your terminal (create environment file .env and run the app!):

cp env.sample .env

python3 app.py or python app.py

• Check if it's running okay. You should see the following output:

Geral-MacBook-Pro:iykra-sentiment-analysis geral$ python3.6 app.py
 * Serving Flask app "app" (lazy loading)
 * Environment: production
   WARNING: This is a development server. Do not use it in a production deployment.
   Use a production WSGI server instead.
 * Debug mode: off
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)

• If it's running okay then hit your API by running this curl command on your terminal:

curl --location --request POST 'http://localhost:5000/classify' \
--header 'Content-Type: application/json' \
--data-raw '{
    "text": "im glad ur doing well"
}'

• You should see this output as the result:

{
  "data": {
    "sentiment": "positive",
    "text": "im glad ur doing well"
  }
}

• You can change the value of the text field with any English sentence to see the predicted sentiment.

• Alternatively, if the above doesn't work for you, you can access the model that I've deployed on Heroku by running this command:

curl --location --request POST 'https://iykra-sentiment-analysis-geral.herokuapp.com/classify' \
--header 'Content-Type: application/json' \
--data-raw '{
    "text": "im glad ur doing well"
}'

• Our goal for this hands on session is to get your model deployed in Heroku just like the above.

Dataset

Please refer here: https://github.com/geraldzakwan/iykra-sentiment-analysis/tree/main/data.

Libraries

Get to know about these two main libraries for our project:

• scikit-learn

• flask

and briefly about these other libraries:

• pandas

• nltk

• gunicorn

Questions

• If you have difficulties or find some errors following the above instructions, please email me (geraldi.dzakwan@gmail.com) with email subject: IYKRA Getting Ready - {Your name}, e.g. IYKRA Getting Ready - Geraldi Dzakwan.

• Explain your difficulties/errors and provide screenshot if any.

• Thanks and see you Friday!

Cheers,

Geral