var shape=[2,3];

var a=tf.tensor([1,2,3,4,4,5],shape);

a.print()

var a=tf.tensor([[1,2,3],[4,4,5]]);

a.print()

var a=tf.scalar(4);

a.print();

var b=tf.tensor1d([0,2,3,4]);

b.print();

var c=tf.tensor2d([[0,3],[4,5]]);

c.print();

tf.zeros([10]).print();
tf.zeros([2,8]).print();
tf.zeros([2,2,3]).print();
tf.ones([10]).print();
tf.ones([2,8]).print();
tf.ones([2,2,3]).print();

var initalValues=tf.ones([8]);

initalValues.print();

//biases±äÁ¿£¬Í¨¹ýassign·½·¨¸üÐÂÆäÖµ

var biases=tf.variable(initalValues);

biases.print();

var updatedValues=tf.tensor1d([0,1,2,3,4,5,6,7]);

updatedValues.print();

biases.assign(updatedValues);

biases.print();

//square

var d=tf.tensor2d([[1,2,3],[4,5,6]]);

d.print();

var d_squared=d.square();

d_squared.print();

//add

var a=tf.tensor2d([[1,2,3],[4,5,6]]);

var b=tf.tensor2d([[3,1,9],[14,25,16]]);

a.print();
b.print();

var c=a.add(b);

c.print();

d.add(b).square().print();

// ¶¨ÒåÒ»¸ö predict º¯Êý

function predict(input) {

// ʵÏÖÒ»¸öÊýѧº¯ÊýµÄ¼ÆËã y = a * x ^ 2 + b * x + c

return tf.tidy(() => {

const x = tf.scalar(input);

const ax2 = a.mul(x.square());

const bx = b.mul(x);

const y = ax2.add(bx).add(c);

return y;
});
}

// ¶¨Òå³£Á¿

var a = tf.scalar(2),
b = tf.scalar(4),
c = tf.scalar(8);

// ²âÊÔÏ predict º¯Êý

predict(1999993).print();

// y = 3 ^ 2 + 1

var y = tf.tidy(() => {

// a, b, ÒÔ¼° one ½«»á±»Çå¿Õµ± tidy ½áÊøʱ¡£

const one = tf.scalar(1);

const a = tf.scalar(3);


const b = a.square();



console.log('tensors µÄÊýÁ¿ (in tidy): ' + tf.memory().numTensors);




return b.add(one);
});

console.log('tensors µÄÊýÁ¿ (outside tidy): ' + tf.memory().numTensors);

y.print();

const x=tf.tensor2d([[0,2,3],[1,2,3]]);

const x_squared=x.square();

x.print();

x_squared.print();

console.log(tf.memory().numTensors);

x.dispose();

console.log(tf.memory().numTensors);

x_squared.dispose();

console.log(tf.memory().numTensors);

http://www.parceljs.io

MobileNets:
Efficient Convolutional Neural Networks for Mobile Vision Applications

https://github.com/fchollet/deep-learning-models/releases/download/v0.6/mobilenet_2_5_224_tf.h5

pip install tensorflowjs

tensorflowjs_converter --input_format keras mobilenet_2_5_224_tf.h5 model

npm install http-server -g

http-server -p 3000 --cors

const model = await tf.loadModel(¡®https://localhost:3000/model.json');

async function loadMobilenet() {

const mobilenet = await tf.loadModel(
'https://storage.googleapis.com/tfjs-models/tfjs/mobilenet_v1_0.25_224/model.json');

console.log(mobilenet.layers)

const layer = mobilenet.getLayer(¡®conv_pw_13_relu');

console.log(layer.output.shape)

return tf.model({
inputs: mobilenet.inputs,
outputs: layer.output
});

}

console.log(layer.output.shape)

model = tf.sequential({

layers: [
tf.layers.flatten({
inputShape: [7, 7, 256]
}),

tf.layers.dense({
units: ui.getDenseUnits(),
activation: 'relu',
kernelInitializer: 'varianceScaling',
useBias: true
}),


tf.layers.dense({
units: NUM_CLASSES,
kernelInitializer: 'varianceScaling',
useBias: false,
activation: 'softmax'
})



]
});

model=tf.sequential();

model.add(tf.layers.flatten({inputShape:[12,4]}));

nn=model.predict(tf.ones([99,12,4]));

console.log(nn.shape);

nn.print()