https://github.com/javascriptdata/scikit.js
JavaScript package for predictive data analysis and machine learning
https://github.com/javascriptdata/scikit.js
Last synced: 9 months ago
JSON representation
JavaScript package for predictive data analysis and machine learning
- Host: GitHub
- URL: https://github.com/javascriptdata/scikit.js
- Owner: javascriptdata
- License: mit
- Created: 2021-09-28T14:55:10.000Z (over 4 years ago)
- Default Branch: main
- Last Pushed: 2023-02-09T00:46:16.000Z (almost 3 years ago)
- Last Synced: 2025-04-04T15:25:54.494Z (10 months ago)
- Language: TypeScript
- Size: 9.24 MB
- Stars: 137
- Watchers: 6
- Forks: 13
- Open Issues: 22
-
Metadata Files:
- Readme: README.md
- Changelog: CHANGELOG.md
- Contributing: CONTRIBUTING_GUIDE.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
# scikit.js
[](https://coveralls.io/github/javascriptdata/scikit.js?branch=main) [](https://github.com/javascriptdata/scikit.js/actions/workflows/release.yml)
TypeScript package for predictive data analysis, data preparation and machine learning.
Aims to be a Typescript port of the [scikit-learn](https://scikit-learn.org) python library.
This library is for users who wish to train or deploy their models to JS environments (browser, mobile) but with a familiar API.
Generic math operations are powered by [Tensorflow.js](https://www.tensorflow.org/js) core layer for faster calculation.
Documentation site: [www.scikitjs.org](https://www.scikitjs.org)
# Installation
### Frontend Users
For use with modern bundlers in a frontend application, simply
```bash
npm i @tensorflow/tfjs scikitjs
```
We depend on the tensorflow library in order to make our calculations fast, but we don't ship it in our bundle.
We use it as a peer dependency. General usage is as follows.
```js
import * as tf from '@tensorflow/tfjs'
import * as sk from 'scikitjs'
sk.setBackend(tf)
```
This allows us to build a library that can be used in Deno, Node, and the browser with the same configuration.
### Backend Users
For Node.js users who wish to bind to the Tensorflow C++ library, simply import the tensorflow C++ version, and use that as the tf library
```bash
npm i @tensorflow/tfjs-node scikitjs
```
```js
const tf = require('@tensorflow/tfjs-node')
const sk = require('scikitjs')
sk.setBackend(tf)
```
Note: If you have ESM enabled (by setting type="module" in your package.json), then you can consume this library with import / export, like in the following code block.
```js
import * as tf from '@tensorflow/tfjs-node'
import * as sk from 'scikitjs'
sk.setBackend(tf)
```
### Script src
For those that wish to use script src tags, simply
```html
import * as tf from 'https://cdn.skypack.dev/@tensorflow/tfjs'
import * as sk from 'https://cdn.skypack.dev/scikitjs'
sk.setBackend(tf)
// or alternatively you can pull the bundle from unpkg
// import * as sk from "https://unpkg.com/scikitjs/dist/web index.min.js"
```
## Simple Example
```js
import * as tf from '@tensorflow/tfjs'
import { setBackend, LinearRegression } from 'scikitjs'
setBackend(tf)
const lr = new LinearRegression({ fitIntercept: false })
const X = [[1], [2]] // 2D Matrix with a single column vector
const y = [10, 20]
await lr.fit(X, y)
lr.predict([[3], [4]]) // roughly [30, 40]
console.log(lr.coef)
console.log(lr.intercept)
```
# Coming from Python?
This library aims to be a drop-in replacement for scikit-learn but for JS environments. There are some
differences in deploy environment and underlying libraries that make for a slightly different experience.
Here are the 3 main differences.
### 1. Class constructors take in objects. Every other function takes in positional arguments.
While I would have liked to make every function identical to the python equivalent, it wasn't possible. In python,
one has named arguments, meaning that all of these are valid function calls.
#### Python
```py
def myAdd(a=0, b=100):
return a+b
print(myAdd()) # 100
print(myAdd(a=10)) # 110
print(myAdd(b=10)) # 10
print(myAdd(b=20, a=20)) # 40 (order doesn't matter)
print(myAdd(50,50)) # 100
```
Javascript doesn't have named parameters, so one must choose between positional arguments, or passing in a single object with all the parameters.
For many classes in scikit-learn, the [constructors take in a ton of arguments](https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier) with sane defaults, and the user usually only specifies which one they'd like to change. This rules out the positional approach.
After a class is created most function calls really only take in 1 or 2 arguments (think fit, predict, etc). In that case, I'd rather simply pass them positionally. So to recap.
#### Python
```py
from sklearn.linear_model import LinearRegression
X, y = [[1],[2]], [10, 20]
lr = LinearRegression(fit_intercept = False)
lr.fit(X, y)
```
Turns into
#### JavaScript
```js
import * as tf from '@tensorflow/tfjs'
import { setBackend, LinearRegression } from 'scikitjs'
setBackend(tf)
let X = [[1], [2]]
let y = [10, 20]
let lr = new LinearRegression({ fitIntercept: false })
await lr.fit(X, y)
```
You'll also notice in the code above, these are actual classes in JS, so you'll need to `new` them.
### 2. underscore_case turns into camelCase
Not a huge change, but every function call and variable name that is `underscore_case` in python will simply be `camelCase` in JS. In cases where there is an underscore but no word after, it is removed.
#### Python
```py
from sklearn.linear_model import LinearRegression
X, y = [[1],[2]], [10, 20]
lr = LinearRegression(fit_intercept = False)
lr.fit(X, y)
print(lr.coef_)
```
Turns into
#### JavaScript
```js
import * as tf from '@tensorflow/tfjs'
import { setBackend, LinearRegression } from 'scikitjs'
setBackend(tf)
let X = [[1], [2]]
let y = [10, 20]
let lr = new LinearRegression({ fitIntercept: false })
await lr.fit(X, y)
console.log(lr.coef)
```
In the code sample above, we see that `fit_intercept` turns into `fitIntercept` (and it's an object). And `coef_` turns into `coef`.
### 3. Always await calls to .fit or .fitPredict
It's common practice in Javascript to not tie up the main thread. Many libraries, including tensorflow.js only give an async "fit" function.
So if we build on top of them our fit functions will be asynchronous. But what happens if we make our own estimator that has a synchronous fit function? Should we burden the user with finding out if their fit function is async or not, and then "awaiting" the proper one? I think not.
I think we should simply await all calls to fit. If you await a synchronous function, it resolves immediately and you are on your merry way. So I literally await all calls to .fit and you should too.
#### Python
```py
from sklearn.linear_model import LogisticRegression
X, y = [[1],[-1]], [1, 0]
lr = LogisticRegression(fit_intercept = False)
lr.fit(X, y)
print(lr.coef_)
```
Turns into
#### JavaScript
```js
import * as tf from '@tensorflow/tfjs'
import { setBackend, LogisticRegression } from 'scikitjs'
setBackend(tf)
let X = [[1], [-1]]
let y = [1, 0]
let lr = new LogisticRegression({ fitIntercept: false })
await lr.fit(X, y)
console.log(lr.coef)
```
## Contribution Guide
See [guide](https://github.com/opensource9ja/scikit.js/blob/main/CONTRIBUTING_GUIDE.md)