This repository contains implementation of second idea challenge by camicroscope for GSOC.
Implement a web based GUI interface that allow user to select a layer for training of deep learning models.
- HTML
- CSS
- Javascript
- Tensorflow js
Nothing. The repository is self-contained. User just have to run index.html file.
Access the page from here .
- The data taken for training is for demonstration purposes and is hard-coded tensorflow tensors. It can be changed by going to the functions addData1() and addData2() for regression and classification respectively in the file main.js in javascript folder.
- Since the data is fixed the input dimension for regression data is 1 and it has 9 sample data point to show that the program is working well.
- Similarly, the input dimension for classification data is 2 it has 10 sample data point and 2 classes to show that program is working well.
- The implementation currently doesn't have any feature for providing testing data which can be added later.