42:null 43:["$","title",null,{"children":"Quick sentiment analysis with TensorFlow | a-k | Observable"}] 49:[] 47:{"id":"ba958d084533bb93","slug":"quick-sentiment-analysis-with-tensor-flow","title":"Quick sentiment analysis with TensorFlow","thumbnail":"3970c9484cc8a19f2df89d664f0f14bb56d1b7b410226c42a25ae93afa30c504","access_level":"public","create_time":"$D2026-08-17T14:39:08.260Z","update_time":"$D2026-08-17T14:39:21.454Z","view_time":null,"first_public_time":"$D2026-08-17T14:39:12.998Z","has_unpublished_edits":null,"fork":{"id":"02c5beb9f86a289c","slug":"quick-sentiment-analysis-with-tensorflow","title":"Quick sentiment analysis with TensorFlow","owner":{"id":"b47c3705cfcd7c1b","login":"tmcw","name":"Tom MacWright"}},"stars":1,"starred":false,"owner":{"type":"team","id":"0d89ff0d43911832","login":"a-kyselova","name":"a-k","avatar_url":"https://avatars-next.observableusercontent.com/c17731a6346b5dfce53e427b1b871e127393f84a1789e18931f02dc51f940fcc","tier":"starter_2024","bio":"","following":false},"version":172,"publish_version":172,"role":null,"authored":false,"authors":[{"type":"individual","id":"4abdd3c5ac6f602a","login":"akyselova","name":"Anastasiia","avatar_url":"https://avatars-next.observableusercontent.com/f400c814810bb16bc0121a8537885d2f84ff777dc43a704664ad27987b43d9df","tier":null,"bio":"","following":false}],"members":[{"type":"team","id":"0d89ff0d43911832","login":"a-kyselova","name":"a-k","avatar_url":"https://avatars-next.observableusercontent.com/c17731a6346b5dfce53e427b1b871e127393f84a1789e18931f02dc51f940fcc","tier":"starter_2024","role":"editor"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":null,"type":"notebook","body":{"titling":"auto","title":"Quick sentiment analysis with TensorFlow","cells":[{"id":0,"value":"md`# Quick sentiment analysis with TensorFlow\n\nBased on [TensorFlow’s example](https://github.com/tensorflow/tfjs-examples/tree/master/sentiment), this is a simple demo of determining positive/negative sentiment with tensorflow. Use the text box below to test it out.`","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":29,"value":"viewof inputText = html``","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":102,"value":"{\n const dim = 100;\n const context = DOM.context2d(dim, dim);\n const radius = dim / 2;\n var arc = d3\n .arc()\n .outerRadius(radius - 1)\n .innerRadius(radius - 34)\n .padAngle(0.1)\n .context(context);\n\n var pie = d3.pie();\n\n var arcs = pie([sentiment, 1 - sentiment]);\n\n context.translate(dim / 2, dim / 2);\n let colors = [\"lightgreen\", \"salmon\"];\n\n context.globalAlpha = 0.5;\n arcs.forEach(function(d, i) {\n context.beginPath();\n arc(d);\n context.fillStyle = colors[i];\n context.fill();\n });\n\n context.globalAlpha = 1;\n context.beginPath();\n arcs.forEach(arc);\n context.lineWidth = 1.5;\n context.stroke();\n\n return html`

${\n context.canvas\n }

\nPositive
\nNegative\n

`;\n}","pinned":false,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":37,"value":"sentiment = {\n const predictOut = model.predict(inputTensor);\n const score = predictOut.dataSync()[0];\n predictOut.dispose();\n return score;\n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":24,"value":"inputTensor = {\n const inputBuffer = tf.buffer([1, metadata.max_len], \"float32\");\n for (let i = 0; i < inputText.length; ++i) {\n const word = inputText[i];\n inputBuffer.set(metadata.word_index[word] + metadata.index_from, 0, i);\n }\n const input = inputBuffer.toTensor();\n return input;\n}","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":14,"value":"model = await tf.loadModel(\n \"https://storage.googleapis.com/tfjs-models/tfjs/sentiment_cnn_v1/model.json\"\n)","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":8,"value":"metadata = (await fetch(\n \"https://storage.googleapis.com/tfjs-models/tfjs/sentiment_cnn_v1/metadata.json\"\n)).json()","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":3,"value":"tf = require(\"@tensorflow/tfjs@0.10.0\")","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"},{"id":100,"value":"d3 = require('d3@5')","pinned":true,"hidden":false,"mode":"ojs","author":"$undefined","summary":"$undefined","tool":"$undefined","providerMetadata":"$undefined","attachments":"$undefined","output":"$undefined","database":"$undefined"}],"files":"$W49","stdlib":"2"}} 48:"$47:body"