2e:null 2f:["$","title",null,{"children":"suggestions:@aman-tiwari/tensorflow-js-weaving-network-drafts | Fabian Iwand | Observable"}] 34:T67e, { sketch; const normalize = arr => tf.tidy(() => { const max = arr.max() const min = arr.min() return arr.sub(min).div(max.sub(min)) }) const woven = tf.tidy(() => { const targetT = tf.image.resizeNearestNeighbor(tf.fromPixels(p5canvas.querySelector('canvas')), [16, 16]) .split([1, 2],2)[0].toFloat().div(255.0).squeeze() const treadleT = tf.tensor(treadling).variable() const tieupT = tf.tensor(tieup).variable() const threadingT = tf.tensor(threading).variable() const forward = () => tf.sigmoid(weave(tf.sigmoid(threadingT.mul(2.0).sub(1.0)), tf.sigmoid(tieupT.mul(2.0).sub(1.0)), tf.sigmoid(treadleT.mul(2.0).sub(1.0)))) const loss = (woven, image) => woven.sub(image).square().sum() // tf.losses.softmaxCrossEntropy(woven, image) const learningRate = 0.01 const optimizer = tf.train.sgd(learningRate) for (let i = 0; i < 10000; i++) { optimizer.minimize(() => loss(forward(), targetT)); } const woven = forward() const resultVis = tf.sigmoid(threadingT.mul(2.0).sub(1.0)).concat(woven) .concat(tf.sigmoid(tieupT.mul(2.0).sub(1.0)) .concat(tf.sigmoid(treadleT.mul(2.0).sub(1.0))), 1).expandDims(2) return tf.image.resizeNearestNeighbor(resultVis, [resultVis.shape[0] * 8, resultVis.shape[1] * 8]) return tf.image.resizeNearestNeighbor(woven.concat(targetT).expandDims(2), [woven.shape[0] * 8 * 2, woven.shape[1] * 8]) }) await tf.toPixels(woven, resultCanvas) tf.dispose([woven]) return 'optimized' }35:[] 32:{"id":"ab2eee099e3efbcc","slug":"suggestions-aman-tiwari-tensorflow-js-weaving-network-dra","title":"suggestions:@aman-tiwari/tensorflow-js-weaving-network-drafts","thumbnail":"78f956e602ddbb07b2465ce2a2c1a2bb428850c357e17d7c1553a04b166a1f24","access_level":"public","create_time":"$D2019-07-10T19:22:12.358Z","update_time":"$D2019-09-21T17:11:51.673Z","view_time":null,"first_public_time":"$D2019-09-21T15:13:56.930Z","has_unpublished_edits":null,"fork":{"id":"85b9564fb4950d3c","slug":"tensorflow-js-weaving-network-drafts","title":"Tensorflow.js Weaving - Optimizing Network Drafts","owner":{"id":"296afa8270178104","login":"aman-tiwari","name":"Aman Tiwari"}},"stars":0,"starred":false,"owner":{"type":"team","id":"898d35f8860c7f03","login":"mootari","name":"Fabian Iwand","avatar_url":"https://avatars-next.observableusercontent.com/0d7defa821f38094c03bad23b9b360a5364e6e97e21fc238c39ddc48db7994ad","tier":"starter_2024"},"version":199,"publish_version":199,"role":null,"authored":false,"authors":[{"type":"individual","id":"07362516b5994994","login":"mootari","name":"Fabian Iwand","avatar_url":"https://avatars-next.observableusercontent.com/0d7defa821f38094c03bad23b9b360a5364e6e97e21fc238c39ddc48db7994ad","tier":null}],"members":[{"type":"individual","id":"07362516b5994994","login":"mootari","name":"Fabian Iwand","avatar_url":"https://avatars-next.observableusercontent.com/0d7defa821f38094c03bad23b9b360a5364e6e97e21fc238c39ddc48db7994ad","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":{"id":"6fb5e804b025ed6d","slug":"suggestions","title":"Suggestions","owner":{"id":"898d35f8860c7f03","login":"mootari","name":"Fabian Iwand"}},"type":"notebook","body":{"title":"suggestions:@aman-tiwari/tensorflow-js-weaving-network-drafts","cells":[{"id":193,"value":"html`\n\n

Fixes: tfjs no longer compatible, lock to version 0.15`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":0,"value":"md`# Tensorflow.js Weaving - Optimizing Network Drafts`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":4,"value":"tf = require('@tensorflow/tfjs@0.15')","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":6,"value":"weave = function weave(threading, tieup, treadling) {\n  return tf.tidy(() => tf.matMul(tf.matMul(treadling, tieup, false, true), threading))\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":8,"value":"canvas = html``","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":10,"value":"treadling = [[1, 0, 0, 0], \n             [0, 1, 0, 0], \n             [0, 0, 1, 0], \n             [0, 0, 0, 1],\n            [1, 0, 0, 0], \n             [0, 1, 0, 0], \n             [0, 0, 1, 0], \n             [0, 0, 0, 1],\n            [1, 0, 0, 0], \n             [0, 1, 0, 0], \n             [0, 0, 1, 0], \n             [0, 0, 0, 1],\n            [1, 0, 0, 0], \n             [0, 1, 0, 0], \n             [0, 0, 1, 0], \n             [0, 0, 0, 1]]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":12,"value":"threading = [[1, 0, 0, 0,  1, 0, 0, 0,  1, 0, 0, 0,  1, 0, 0, 0],\n             [0, 1, 0, 0,  0, 1, 0, 0,  0, 1, 0, 0,  0, 1, 0, 0],\n             [0, 0, 1, 0,  0, 0, 1, 0,  0, 0, 1, 0,  0, 0, 1, 0],\n             [0, 0, 0, 1,  0, 0, 0, 1,  0, 0, 0, 1,  0, 0, 0, 1]]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":17,"value":"tieup = [[1, 1, 0, 0],\n         [0, 1, 0, 0],\n         [0, 0, 1, 0],\n         [0, 0, 0, 1]]","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":19,"value":"{\n  const result = tf.tidy(() => {\n    const woven = weave(threading, tieup, treadling)\n    return tf.image.resizeNearestNeighbor(woven.expandDims(2), [woven.shape[0] * 8, woven.shape[1] * 8])\n  })                     \n  await tf.toPixels(result, canvas)\n  result.dispose()\n  return 'weaving...'\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":82,"value":"resultCanvas = html``","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":69,"value":"$34","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":26,"value":"{ now; yield tf.memory() }","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":148,"value":"f = 1","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":150,"value":"g = f * 2","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":38,"value":"p5 = require('p5')","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":98,"value":"viewof clear = html``","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":40,"value":"sketch = function (p) {\n  clear;\n  p.setup = function() { p.createCanvas(256, 256); p.background('white'); p.noStroke(); p.fill('black')  }\n  p.draw = function() { }\n  p.mouseDragged = function() {\n    p.rect(p.mouseX, p.mouseY, 10, 10) \n  }\n  p.keyPressed = function() {\n    if(p.key == 'c') p.background('white') \n  }\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":43,"value":"p5canvas = {\n  sketch;\n  return html``\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"},{"id":45,"value":"p5sketch = new p5(sketch, p5canvas)","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W35","stdlib":"1","titling":"auto"}}
33:"$32:body"