Flask API 🌐 has several routes that allow you to make predictions, ingest data, and retrain the model.
This repository contains two main folders: data and ejercicio. The data folder contains the necessary data for the Advertising Model API, while the ejercicio folder contains the code for the API and the script to convert a CSV file into an SQLite database.
Advertising Model API 🚀 This repository contains a Flask API for an advertising model and a script to convert a CSV file into an SQLite database.
Flask API 🌐 The Flask API has several routes that allow you to make predictions, ingest data, and retrain the model.
Routes 🛣️ 🌈 The / route accepts GET requests and returns a warm welcome message. 🤗
🔮 The /v2/predict route accepts GET requests with values for tv, radio, and newspaper as arguments in the call. When a request is received, the code loads a previously trained model from the data/advertising_model file, gets the values of the arguments, and uses the model to make a prediction. Then, it returns the prediction to the client. 🔮
📬 The /v2/ingest_data route accepts POST requests with data in JSON format in the request body. When a request is received, the code connects to an SQLite database called advertising.db in the data directory, creates a table called advertising if it doesn’t exist, and stores the received data in this table. 📬
🔧 The /v2/retrain route accepts PUT requests. When a request is received, the code loads the previously trained model from the data/advertising_model file, connects to the SQLite database, retrieves the data stored in the advertising table, converts it into a pandas DataFrame, and separates the independent and dependent variables. Then, it retrains the model using this data and saves the retrained model in the same file. 🔧
SQL Database Creator Script 🗄️ The sql_db_creator.py script reads data from a CSV file called Advertising.csv in the data directory, connects to an SQLite database called advertising.db in the same directory, and stores the data from the CSV file in this database.
The script uses pandas to read the CSV file and convert it into a DataFrame. Then, it renames the column newspaper to newpaper, as specified in your requirements. Finally, it uses the to_sql method of pandas to create a table called advertising in the SQLite database and store the data from the DataFrame in this table.
Usage 💻 To use this repository, first make sure you have Python 3 installed on your machine. Then, clone this repository and navigate to its root directory.
To run the Flask API, use the following command:
python app_model_db.py
This will start a local development server at http://localhost:5000. You can then use your web browser or an HTTP client like cURL or Postman to interact with the API.
To run the SQL database creator script, use the following command:
- python csv_to_sqlite.py
This will read data from the CSV file and store it in an SQLite database called advertising.db in the data directory.