https://github.com/cihaneksi/salary_regression

In this project I trained a machine learning model to predict customer estimated salary based on various features. Solution method is Artificial Neural Network (ANN).
https://github.com/cihaneksi/salary_regression

ai ann keras prediction python regression tensorflow

Last synced: 2 months ago
JSON representation

In this project I trained a machine learning model to predict customer estimated salary based on various features. Solution method is Artificial Neural Network (ANN).

• Host: GitHub
• URL: https://github.com/cihaneksi/salary_regression
• Owner: CihanEksi
• Created: 2025-02-20T22:20:27.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-02-20T22:22:14.000Z (over 1 year ago)
• Last Synced: 2025-02-20T23:24:42.404Z (over 1 year ago)
• Topics: ai, ann, keras, prediction, python, regression, tensorflow
• Language: Python
• Homepage:
• Size: 290 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Salary Regression Model

In this project I trained a machine learning model to predict customer estimated salary based on various features.

Solution method is Artificial Neural Network (ANN).

## Requirements

- Python 3.x

- TensorFlow

- pandas

- pickle

- scikit-learn

## Files

- `regression_model.keras`: Trained TensorFlow model

- `label_encoder.pkl`: Pickle file for categorical encoding

- `ordinal_encode.pkl`: Pickle file for ordinal encoding

- `scaler.pkl`: Pickle file for feature scaling

## Prediction Steps

1. **Data Preprocessing**

   - Load the model, encoders, and scaler

   - Convert input data to pandas DataFrame

   - Apply categorical encoding for the Gender field

   - Apply ordinal encoding for the Geography field

   - Scale the features using the scaler

2. **Model Prediction**

   - Feed the preprocessed data into the model

   - Get the prediction output

   - Return the prediction as a JSON response

## Input Features

- `CreditScore`: Customer's credit score (numeric)

- `Geography`: Customer's location (France, Spain, Germany)

- `Gender`: Customer's gender (Male, Female)

- `Age`: Customer's age (numeric)

- `Tenure`: Number of years as a customer (numeric)

- `Balance`: Account balance (numeric)

- `NumOfProducts`: Number of bank products used (numeric)

- `HasCrCard`: Has credit card (1 = Yes, 0 = No)

- `IsActiveMember`: Active member status (1 = Yes, 0 = No)

- `Exited`: Customer churn status (1 = Yes, 0 = No)

## What We Did?

1. **Loading Resources**

   - Loaded the trained model using `load_model`.

   - Loaded the encoders and scaler from their respective pickle files.

2. **Data Encoding**

   - Defined functions for ordinal and categorical encoding.

   - Encoded the input data using these functions.

3. **Data Scaling**

   - Scaled the encoded input data using the loaded scaler.

4. **Prediction**

   - Made predictions using the preprocessed and scaled data.

   - Converted the prediction result to a JSON format for easy interpretation.