Parsing NBA Statistics

Parsing NBA Statistics

---

Table Of Contents

• Overview
• Project Structure
• Instructions On Running The Project
• Process Descriptions

Overview

In this project, we parse nba data, load into a database and calculate some statistics from our db. Reading in the data from our normalized database we also predict the result of a match between two opponents. Data is made accesible via a command line interface.

As as defined below:

• Parse NBA statistics provided in the archive files
• Dump the statistics into a MySQL database in a normalized format
• Create a user facing functionality to retrieve the following data points:
• The best player in terms of productivity for each week of the selected season
• Prediction of a match result between two teams
• Exterior command line interface

Project Structure

Project Main Folder |--start_app.py #runs the main app on the command line
|--player_efficiency.py #contains functions run player efficiency by week
|
|--data
| |--init.py #module import | |--dbfile #will hold the database when initiated
| |--archive.zip #zipped file of data to process
| |--create_db.py #python file with functions for creating a db instance
| |--process_data.py #python file for data processing & cleaning
| |--process_dataframes.py #python file for additional data processing & cleaning
|
|--models
| |--init.py #module import | |--classifier.pkl.csv #will hold the classifier of the ml model
| |--train_classifier.py #python script to train model on data
|
|--README.md
|--requirements.txt

Instructions On Running The Project

Environment Optional: If you plan to use a virtual environment. In the terminal run the following commands: - python3 -m venv environment-name
- Follow up by running:
- Windows: run venv\Scripts\activate || On Linux/Mac:source environment-name/bin/activate
- python3 -m pip install --upgrade pip
- pip3 install -r requirements.txt
- To deactivate: deactivate

Compulsory Setup

1. Run the following commands in the project's root directory to set up the database and classifier model.

   • To run ETL pipeline that cleans data and stores in database python3 data/process_data.py data/archive.zip data/mydb.db

   • To run ML pipeline that trains classifier and saves to disk python3 models/train_classifier.py data/mydb.db models/classifier.pkl

2. Run the following command in the app's directory to run the terminal app. python3 start.py data/mydb.db models/classifier.pkl

Process Descriptions

The project can be separted into three sections, each with their contributions to the application.

1. ETL Pipeline A Python script, process_data.py and create_db.py, that runs a data cleaning pipeline that:

• Create the db with create_db.py
• Loads the archive datasets
• Merges the two datasets
• Cleans the data
• Stores it in a SQLite database

2. ML Pipeline In a Python script, train_classifier.py, that runs a machine learning pipeline that:

• Loads data from the SQLite database
• Splits the dataset into training and test sets
• Builds a text processing and machine learning pipeline
• Trains and tunes a model
• Exports the final model as a pickle file

3. ML Pipeline In a Python script, player_efficiency.py, calculates stats from the database: Productivity is defined as: Efficiency in Basketball Wiki (PTS + REB + AST + STL + BLK − (Missed_FG + Missed_FT + TO)) / GP

• Calculates the player productivity statistics via an sqlalchemy orm
• Prints the db to the command line
• Exports the final model as a pickle file

4. Command Line Application A small interactive command line application that enables you in getting stats and data from the Database