https://github.com/ehsangazar/linear-machine-learning-project
Linear Machine Learning Project using Tensorflow and React
https://github.com/ehsangazar/linear-machine-learning-project
Last synced: 7 months ago
JSON representation
Linear Machine Learning Project using Tensorflow and React
- Host: GitHub
- URL: https://github.com/ehsangazar/linear-machine-learning-project
- Owner: ehsangazar
- Created: 2018-05-29T03:13:01.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-05-29T03:40:30.000Z (over 7 years ago)
- Last Synced: 2025-01-23T11:45:47.390Z (9 months ago)
- Language: JavaScript
- Homepage: https://ehsangazar.com/linear-machine-learning-and-tensorflowjs-5e212b8a318d
- Size: 1.19 MB
- Stars: 1
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Linear machine learning project
I have started learning machine learning and this is the basic of tensorflow.js and linear learning algorithm
Demo: https://ehsangazar.github.io/linear-machine-learning-project/
## Importing tensorflow
```
import * as tf from '@tensorflow/tfjs';
```
## Initializing the Model
- Sequential because it will be a sequence of numbers
- dense layers are fully connected layers
- units is dimensionality of the output space
- inputShape will be used to create an input layer to insert before this layer
- sgd: Stochastic gradient descent
https://en.wikipedia.org/wiki/Stochastic_gradient_descent
- meanSquaredError: this is our loss definition
```
this.linearModel = tf.sequential();
// linear
this.linearModel.add(tf.layers.dense({
units: 1,
inputShape: [1]
}))
// preate the model for training with defnining Loss function
this.linearModel.compile({
optimizer: 'sgd',
loss: 'meanSquaredError'
});
```
## Training with Data
Two arrays for using as a training data
```
const xs = tf.tensor1d([1.2, 2.4, 3.5, 4.71, 5.98, 6.168, 7.779, 8.182, 9.59, 2.16, 7.042, 10.71, 5.313, 7.97, 5.654, 9.7, 3.11]);
const ys = tf.tensor1d([4.6, 5.7, 6.9, 9.19, 10.684, 12.53, 23.366, 32.596, 42.53, 1.22, 2.87, 3.45, 1.65, 2.904, 2.42, 2.4, 1.31]);
this.linearModel.fit(xs, ys)
```
## Prediction for the next number
This will predict a number based on a value
```
this.linearModel.predict(tf.tensor2d([value], [1, 1]))
```
If you want, create a pull-request or just contact me on @ehsangazar