Spaces:
Running
Running
File size: 3,954 Bytes
30e9731 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 |
/* Copyright 2021 Google LLC. All Rights Reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
==============================================================================*/
window.initSent = async function(sent, sel){
var isHamlet = sent.class == 'hamlet'
var isMobile = innerWidth < 900
var sel = d3.select('.' + sent.class)
.st({opacity: .5, marginBottom: isHamlet ? '' : 40})
// Load completitions
var str = sent.str
while (str.includes('__')) str = str.replace('__', '_')
str = str.replace('_', 'things')
var tokens = tokenizer.tokenizeCLS(str)
.filter(d => d < 30522)
var topTokens = await post('embed_group_top', {tokens})
topTokens.forEach(sent => {
sent.forEach(d => d.str = tokenizer.vocab[d.i])
})
var displayTokens = tokens
.slice(1)
.map((vocabIndex, i) => {
return {i, str: bertLargeVocab[vocabIndex].replace('##', '')}
})
displayTokens.pop()
sel.html('').st({opacity: 1})
if (!sel.node()) return
var divSel = sel.append('div')
.st({position: 'relative'})
var svgSel = divSel.append('svg')
.st({position: 'absolute', top: 0, zIndex: -10})
var tokenSel = divSel
.append('div.token-container')
.st({padding: 20, paddingLeft: 0, paddingRight: 0, fontSize: 20})
.appendMany('button.token', displayTokens)
.text(d => d.str)
.on('click', drawToken)
var connectionPath = svgSel.append('path').at({fill: 'none', stroke: '#000', strokeWidth: 1})
var padding = 5
var width = divSel.node().offsetWidth
var botWidth = isMobile ? width - padding*2 : 580
var botTextSel = divSel.append('div.top-sents')
.translate([width/2 - botWidth/2 - padding + .5, 15])
.st({
width: botWidth,
height: 170,
outline: '1px solid #000',
padding: padding,
// position: 'absolute',
background: '#fff',
overflowY: 'scroll',
fontSize: isMobile ? 10 : '',
})
if (isHamlet){
divSel.append('div.caption')
.text(`BERT's predictions for what should fill in the hidden word`)
.st({fontWeight: '', lineHeight: '1.1em', fontSize: 14, textAlign: 'center', width: '100%', marginTop: 20})
}
var curIndex = -1
function drawToken(token){
var node = tokenSel.filter(d => d == token).node()
var x = node.offsetLeft + node.offsetWidth/2
var y = node.offsetTop + node.offsetHeight
var y1 = botTextSel.node().offsetTop
connectionPath.at({d: ['M', x, y, 'L', width/2, y1 + 15].join(' ')})
var completionSel = botTextSel.html('').appendMany('span', topTokens[token.i + 1])
.st({display: 'inline-block', fontFamily: 'monospace', width: isMobile ? '47%' : '31%', borderBottom: '1px solid #ccc', margin: 4, fontSize: innerWidth < 350 ? 12 : isMobile ? 13 : 14 })
completionSel.append('span')
.st({color: '#ccc'})
.html(d => {
var str = d3.format('.3f')(d.p*100) + '% '
if (str.length < 8) str = ' ' + str
return str
})
completionSel.append('span')
.text(d => d.str.replace('▁', ''))
tokenSel
.text(d => d.str)
.classed('active', false)
.filter(d => d == token)
.classed('active', true)
.text(d => d.str.split('').map(d => '_').join(''))
}
var i = displayTokens.length - (isHamlet ? 2 : 2)
if (tokens.includes(2477)) i = tokens.indexOf(2477) - 1
drawToken(displayTokens[i])
var topTokensSel = sel.append('div.top-tokens')
}
if (window.init) init()
|