https://github.com/ekojs/machine_learning
Berisi kumpulan algoritma dari Machine Learning dan Computational Intelligance
https://github.com/ekojs/machine_learning
artificial-intelligence computational-intelligence machine-learning
Last synced: about 1 year ago
JSON representation
Berisi kumpulan algoritma dari Machine Learning dan Computational Intelligance
- Host: GitHub
- URL: https://github.com/ekojs/machine_learning
- Owner: ekojs
- License: mit
- Created: 2017-03-04T12:19:17.000Z (over 9 years ago)
- Default Branch: master
- Last Pushed: 2019-03-21T00:39:27.000Z (over 7 years ago)
- Last Synced: 2025-04-26T19:39:29.772Z (about 1 year ago)
- Topics: artificial-intelligence, computational-intelligence, machine-learning
- Language: JavaScript
- Size: 23.4 KB
- Stars: 4
- Watchers: 1
- Forks: 7
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
### Machine Learning & Computational Intelligence
#### Berisi kumpulan algoritma dari Machine Learning dan Computational Intelligance
### 1. K-Means Clustering Algorithm
Contoh menggunakan nodejs console :
```javascript
var ejs_kmeans = require('./unsupervised/ejs_kmeans');
function TestData(samples,centroid){
console.log('Samples Data : %s \n','('+samples.join(') (')+')');
var k_means = new ejs_kmeans.k_mean_cluster(samples);
k_means.initialize(centroid);
k_means.calculate();
console.log(k_means.result().replace(/
/g,"\n").replace(/ /g,' ').replace(/<\/?strong>/g,''));
}
TestData([[5.09,5.80], [3.24,5.90], [1.68,4.90], [1.00,3.17], [1.48,1.38], [2.91,0.20], [4.76,0.10], [6.32,1.10], [7.00,2.83], [6.52,4.62]],[[1.48,1.38],[4.76,0.10]]);
//TestData([[5.09,5.80], [3.24,5.90], [1.68,4.90], [1.00,3.17], [1.48,1.38], [2.91,0.20], [4.76,0.10], [6.32,1.10], [7.00,2.83], [6.52,4.62]],[[5.09,5.80], [3.24,5.90]]);
//TestData([[1.0,1.0],[1.5,2.0],[3.0,4.0],[5.0,7.0],[3.5,5.0],[4.5,5.0],[3.5,4.5]],[[1,1],[5,7]]);
//TestData([[1,1],[2,1],[4,3],[5,4]],[[1,1],[2,1]]);
//TestData([[1,1,2],[2,1,3],[4,3,2],[5,4,4],[4,4,4]],[[1,1,2],[2,1,3]]);
//TestData([[5.09,5.80], [3.24,5.90], [1.68,4.90], [1.00,3.17], [1.48,1.38], [2.91,0.20], [4.76,0.10], [6.32,1.10], [7.00,2.83], [6.52,4.62]],[[5.09,5.80], [3.24,5.90], [1.68,4.90]]);
```
Baca artikel pada [`K-Means Clustering Algorithm`](http://ekojunaidisalam.com/2017/02/09/k-means-clustering-algorithm/)
Simulasi program di [`sini`](https://ekojs.github.io/ejs_k-means/)
### 2. Backpropagation Neural Network
Penggunaan Backpropagation disini, penulis **masih** menggunakan fungsi aktivasi secara **hardcode**, fungsi aktivasi pada IH menggunakan `TANH` sedangkan pada HO menggunakan fungsi aktivasi `SOFTMAX` dan 'SIGMOID` untuk **Binary Classification**.
Contoh menggunakan nodejs console :
```javascript
var ejs_neural = require('./supervised/ejs_neural');
var opt = {
learning_rate:0.05,
momentum:0.01,
maxEpochs:500
};
// var TT = [[-1,-1,1,-1],[-1,1,1,1],[1,-1,1,1],[1,1,1,-1]]; // xor
// var TT = [[-1,-1,1,-1],[-1,1,1,-1],[1,-1,1,-1],[1,1,1,1]]; // and
var TT = [[-1,-1,1,-1],[-1,1,1,1],[1,-1,1,1],[1,1,1,1]]; // or
// var json = '{"0":-1.8570453538530167,"1":-1.0578781246699556,"2":1.6716922611300928,"3":-1.1001127357871128,"4":1.8538154659419666,"5":-1.124630964477959,"6":-1.677438527019374,"7":-1.0075326202144037,"8":-1.0230813212370369,"9":0.46272458742612566,"10":-0.6512284898272416,"11":-0.6612252388250472,"12":-0.9230821882924356,"13":0.7306611846866936,"14":-1.066699519330741,"15":-0.5033335760896673,"16":3.412458805650443,"17":1.6494378062956072,"18":3.3143332296912598,"19":-1.6840062015951618,"20":0.2616096437809812}';
var nn = new ejs_neural.Neural(3,4,1,opt);
// tes.fromJSON(json);
// nn.initWeights(ejs_neural.range(0.01,3*4+4+4*1+1,0.01))
nn.train(TT);
// var data = nn.makeTrainTest(TT);
// console.log(data);
// console.log('Mulai training ...');
// nn.train(data.data_training);
// console.log('Training selesai ...');
// console.log('--------------- Final nn weights -----------------');
// console.log(nn.toJSON());
// console.log('--------------------------------------------------');
// console.log('Akurasi data training : %d',nn.accuracy(data.data_training));
// console.log('Akurasi data test : %d',nn.accuracy(data.data_test));
console.log('Akurasi data : %d',nn.accuracy(TT));
console.log('Actual %d, prediction : %d',-1,nn.forward([-1,-1,1]).outputs[0]); // -1 or
console.log('Actual %d, prediction : %d',1,nn.forward([1,1,1]).outputs[0]); // 1 or
console.log('Actual %d, prediction : %d',1,nn.forward([-1,1,1]).outputs[0]); // 1 or
console.log('Actual %d, prediction : %d',1,nn.forward([1,-1,1]).outputs[0]); // 1 or
```