https://github.com/shikharvashistha/two-layer-n-n

Two layer neural network for classification and regression problem
https://github.com/shikharvashistha/two-layer-n-n

Last synced: 13 days ago
JSON representation

Two layer neural network for classification and regression problem

• Host: GitHub
• URL: https://github.com/shikharvashistha/two-layer-n-n
• Owner: shikharvashistha
• License: mit
• Created: 2023-08-18T05:08:21.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-08-21T05:13:20.000Z (over 1 year ago)
• Last Synced: 2024-11-13T16:18:17.677Z (2 months ago)
• Language: Python
• Size: 145 KB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
## How to run

- ``` pip install -r requirements.txt ```

- ``` python concrete.py ```

## Specification

- This code has one hidden layer with 50 neurons and iterates 1000 times to find the regression mean squared error and classification accuracy in case of regression and classification respectively.

- It also outputs regression prediction and class in case of regression and classification problem respectively.

- It also outputs the ROC curve for classification problem.

- It also outputs the scatter plot for regression problem.

**Note : Code takes time to run, therefore added the screenshot of output and plots**

## Scatter Plot for regression

![alt text](assets/scatter_plot.png)

## ROC Curve for classification

![alt text](assets/roc_curve.png)

## Screenshot of output

![alt text](assets/hw.png)