An implementation of Tensorflow JS that will try to guess a correct Y value for a linear equation you build.
Build a linear equation in the form y = ax + b.
Other parameters:
| Parameter | Description |
|---|---|
| Sample points | How many (x,y) pairs are generated from your equation and fed to the algorithm |
| Epochs | How many "generations" of the algorithm will be trained on the data before answering |
| X to predict | The x value you want the algorithm to predict the Y value for |
Machine learning in your browser! This uses Tensorflow JS to implement a simple neural network.
Scripts are written for a *NIX environment
Assuming you use yarn:
yarn
yarn watchDefault serving location is localhost:1234
- Better UI
- Graphs of data or Tensorboard
- Support higher order functions