2a:null 2b:["$","title",null,{"children":"IEEEVIS 2023 Papers | John Alexis Guerra Gómez | Observable"}] 31:T475,attrs = [ "paper_type_name", "title", "abstract", // "UID", "authors", "award", // "event_id", "event_title", // "external_paper_link", // "has_image", // "has_pdf", // "id", "image_caption", "keywords", "paper_type", // "paper_type_color", // "prerecorded_video_id", // "prerecorded_video_link", // "session_bunny_ff_link", // "session_bunny_ff_subtitles", // "session_bunny_prerecorded_link", // "session_bunny_prerecorded_subtitles", // "session_id", // "session_room", "session_title", // "session_uid", // "session_youtube_ff_id", // "session_youtube_ff_link", // "session_youtube_prerecorded_id", // "session_youtube_prerecorded_link", // "sessions", "time_stamp", // "youtube_ff_id", // "youtube_ff_url", "time_stamp_parsed", // "thumb", // "slot_id", "contributors", // "uid", "time_start", // "time_end", // "doi", // "fno", "paper_award", // "youtube_ff_link", // "bunny_ff_link", // "bunny_ff_subtitles", // "youtube_prerecorded_link", // "youtube_prerecorded_id", // "bunny_prerecorded_link", // "bunny_prerecorded_subtitles" ]32:T860,create_parameter_form = (dr_method) => { switch (dr_method) { case "ISOMAP": case "LLE": case "LTSA": return html``; break; case "SAMMON": return html``; break; case "TSNE": return html``; break; case "UMAP": return html``; break; case "TriMap": return html``; break; case "LSP": return html``; break; case "LDA": return html`

class labels `; break; default: return html``; break; } }33:T419,// // Plot viz // projection = { // const projection = await Plot.plot({ // marginTop: 10, // inset: 10, // color: { // // scheme: "Paired", // legend: true, // columns: 3 // }, // legend: { // marginTop: 50 // }, // width: width, // height: 400, // x: {axis: null}, // y: {axis: null}, // marks: [ // Plot.dot(dataToPlot, { // x: "x", // y: "y", // fill: colorBy, // r: 5, // opacity: 0.8 // // strokeWidth: 4, // // stroke: "Category" // }), // Plot.tip( // dataToPlot, // Plot.pointer({ // stroke: "paper_type", // // channels: { Note: "Note", sport: "sport" }, // x: "x", // y: "y", // title: embeddingAccessor // }) // ) // ] // }); // Array.from(projection.querySelectorAll("svg")).map(ele => ele.style.overflow="visible"); // console.log( projection.querySelectorAll("svg")); // return projection; // }34:[{"id":"99e1632e48b7ad10cfdd0545508cdeb1d66d0ec951443e6690302a650d4c2bd4c68e3da3239c50aa11a6daffe217f8a2ccd3b83598798661a5dba873d33e36f7","href":"https://static.observableusercontent.com/files/99e1632e48b7ad10cfdd0545508cdeb1d66d0ec951443e6690302a650d4c2bd4c68e3da3239c50aa11a6daffe217f8a2ccd3b83598798661a5dba873d33e36f7","name":"paper_list_ieeevis_2023_from_github@1.json","type":"application/json","lastModified":1698000219640.739,"size":1265218},{"id":"16818c37fa5e316943120bbf921730d01f5d248a0992c69e7fa2eeb4fef673679b5acdfd024eae7b944a6701433afa14e3ad682942239b38fa84ea5387a7fd2a","href":"https://static.observableusercontent.com/files/16818c37fa5e316943120bbf921730d01f5d248a0992c69e7fa2eeb4fef673679b5acdfd024eae7b944a6701433afa14e3ad682942239b38fa84ea5387a7fd2a","name":"papers_vis2023@2.json","type":"application/json","lastModified":1698001289861.287,"size":930002},{"id":"d8166fb7a866d6759bb7b74237f1e7872a543bf456e79965c1c6ca9d4cd286c3c4085bae57daf1535969978acfa8c068abc256e3844790cb455e2aa5a67d0935","href":"https://static.observableusercontent.com/files/d8166fb7a866d6759bb7b74237f1e7872a543bf456e79965c1c6ca9d4cd286c3c4085bae57daf1535969978acfa8c068abc256e3844790cb455e2aa5a67d0935","name":"embeddings@2.json","type":"application/json","lastModified":1698005198254.917,"size":4044935},{"id":"47c556584c730f9cbd92dbb8b575695ed93223eea46714dd876d688b2d2a7306612e3b99079885e53a373030cd59ec486c46181160583fcf92cd8521acc496ca","href":"https://static.observableusercontent.com/files/47c556584c730f9cbd92dbb8b575695ed93223eea46714dd876d688b2d2a7306612e3b99079885e53a373030cd59ec486c46181160583fcf92cd8521acc496ca","name":"embeddings_v2.json","type":"application/json","lastModified":1698021920909.635,"size":4045226},{"id":"97ecc34124be5b9ef9d19374a25781224386a5e42e2adef56c9af8215e40bf44e137f08ee08acaa9e85d814923054c4ae402fb8193a6a2c3802d8fffcdc01205","href":"https://static.observableusercontent.com/files/97ecc34124be5b9ef9d19374a25781224386a5e42e2adef56c9af8215e40bf44e137f08ee08acaa9e85d814923054c4ae402fb8193a6a2c3802d8fffcdc01205","name":"embedding_clusters.json","type":"application/json","lastModified":1698127616554.881,"size":809}] 2e:{"id":"98927c089ccb6347","slug":"ieeevis-2023-papers","title":"IEEEVIS 2023 Papers","thumbnail":"7f4eeced5827321704ab85d002d6d09e0e4acefb5f43c392491afc752b701629","access_level":"public","create_time":"$D2023-10-21T23:17:47.845Z","update_time":"$D2024-04-30T22:23:19.391Z","view_time":null,"first_public_time":"$D2023-10-23T05:59:41.981Z","fork":false,"stars":28,"starred":false,"owner":{"type":"team","id":"cd3c1e49fc84d010","login":"john-guerra","name":"John Alexis Guerra Gómez","avatar_url":"https://avatars-next.observableusercontent.com/21638cee44bb0ea70798da725b57111a5d3e2acebcd14524565992f7d1594aed","tier":"starter_2024"},"version":1651,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"be0076fec73aa550","login":"john-guerra","name":"John Alexis Guerra Gómez","avatar_url":"https://avatars-next.observableusercontent.com/21638cee44bb0ea70798da725b57111a5d3e2acebcd14524565992f7d1594aed","tier":null}],"members":[{"type":"individual","id":"be0076fec73aa550","login":"john-guerra","name":"John Alexis Guerra Gómez","avatar_url":"https://avatars-next.observableusercontent.com/21638cee44bb0ea70798da725b57111a5d3e2acebcd14524565992f7d1594aed","tier":null,"role":"editor"}],"comments":[],"custom_thumbnail":true,"trashed":false,"subscribed":null,"forks":5,"collection":null,"type":"notebook","body":{"title":"IEEEVIS 2023 Papers","cells":[{"id":0,"value":"# IEEEVIS 2023 Papers\n\nThe IEEEVIS 2023 Conference papers organized by their text (title, type and abstract) similarity using sentence embeddings via [HuggingFace Transformer.js](https://observablehq.com/@huggingface/sentence-embeddings-and-dimension-reduction-in-the-browse) and dimensionality reduction methods using [Druid.js](https://observablehq.com/@fil/druidjs-worker).\n\nYou can filter for specific paper tracks and, interact with the scatterplot to explore the different papers.\n\nCredits: [Hugging Face and @Radames](https://observablehq.com/@huggingface/sentence-embeddings-and-dimension-reduction-in-the-browse), [Fil](https://observablehq.com/@fil/), [Rene Cutura](https://observablehq.com/@saehrimnir/hello-druidjs), [Vega-Lite team](https://vega.github.io/vega-lite-api/), [@mootari](https://observablehq.com/@mootari/displaying-progress)","pinned":false,"hidden":false,"mode":"md"},{"id":24,"value":"viewof facetedSelected = conditionalShow(FacetedSearch(papers, {\n attribs: [\n \"session_title\",\n \"paper_type_name\",\n \"event_title\",\n // \"sessions\",\n // \"award\",\n \"title\",\n // \"authors\",\n \"session_room\",\n \"time_stamp_parsed\"\n ]\n}), {label: \"Filter\"})","pinned":false,"hidden":false,"mode":"ojs"},{"id":109,"value":"viewof druid_method = Inputs.select(\n [\n \"UMAP\",\n \"TSNE\",\n \"TriMap\",\n \"PCA\",\n // \"LLE\",\n \"LTSA\",\n \"ISOMAP\",\n \"FASTMAP\",\n \"MDS\",\n \"LSP\",\n // \"LDA\",\n \"TopoMap\",\n \"SAMMON\"\n ].sort(d3.ascending),\n {\n label: \"Dimensionality Reduction Method.\",\n // value: \"UMAP\"\n value: \"TriMap\"\n }\n)","pinned":false,"hidden":false,"mode":"ojs"},{"id":110,"value":"viewof reducerParams = form(create_parameter_form(druid_method))","pinned":false,"hidden":false,"mode":"ojs"},{"id":519,"value":"viewof colorBy = Inputs.select(\n [\n \"paper_type_name\",\n // \"sessions\",\n \"event_title\",\n \"award\",\n // \"title\",\n // \"authors\",\n \"session_room\",\n \"session_title\",\n // \"time_stamp_parsed\"\n ],\n { label: \"Color by\", value: \"paper_type_name\" }\n)","pinned":false,"hidden":false,"mode":"ojs"},{"id":1137,"value":"viewof interactive = Inputs.toggle({\n label: htl.html`Make chart interactive.
Disable to speed up animation`,\n value: true\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":680,"value":"viewof drSelection = interactive\n ? vegaSelected(vegaScatterPlot.toSpec(), { renderer: \"canvas\" })\n : vegaScatterPlot.render()","pinned":false,"hidden":false,"mode":"ojs"},{"id":752,"value":"htl.html` \n \n

Papers selected

\n \n ${papersHighlighted.slice(0, 20).map(\n (p) => htl.html`\n\n

\n ${p.title}\n

\n ${\n p.session_youtube_ff_link ? \n htl.html`⏩ Fast forward ` \n : \"\"\n }\n ${\n p.session_youtube_prerecorded_link ? \n htl.html`📼 Prerecorded talk ` \n : \"\"\n }\n

\n ${\n p.has_image\n ? htl.html`

$\"${p.title}\"$

`\n : \"\"\n }\n \n

${p.abstract}

\n`\n )}\n`","pinned":false,"hidden":false,"mode":"ojs"},{"id":1276,"value":"## Embeddings\n\nHere is a visualization of all the embeddings using navio","pinned":false,"hidden":false,"mode":"md"},{"id":1279,"value":"navio(embeddings.map(d => d.data), {attribWidth: 2, height: 600})","pinned":true,"hidden":false,"mode":"ojs"},{"id":1638,"value":"embeddings","pinned":true,"hidden":false,"mode":"ojs"},{"id":588,"value":"viewof selected = conditionalShow(await navio(facetedSelected), {\n label: \"Filter with Navio\"\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":531,"value":"## Compute Embeddings dynamically\n\nYou can compute the embeddings in the browser if you want","pinned":false,"hidden":false,"mode":"md"},{"id":431,"value":"viewof computeEmbeddings = Inputs.toggle({label: \"Compute Embeddings\"})","pinned":false,"hidden":false,"mode":"ojs"},{"id":544,"value":"viewof computeEmbeddingsUsing = Inputs.checkbox(attrs, {value: [\"title\", \"paper_type_name\", \"abstract\"], label: \"Compute Embeddings using\"})","pinned":false,"hidden":false,"mode":"ojs"},{"id":409,"value":"computeEmbeddings && DOM.download(new Blob(\n [JSON.stringify(embeddings_computed)], \n {type: \"application/json\"}\n ), \"embeddings.json\", \"Download Embeddings\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":261,"value":"viewof restart = computeEmbeddings ? html`` : pleaseEnableComputeEmbeddings()","pinned":false,"hidden":false,"mode":"ojs"},{"id":259,"value":"viewof embeddings_computed = {\n if (computeEmbeddings) {\n const before = performance.now();\n restart;\n const values = [];\n const bar = progress({ interval: 50, invalidation });\n yield bar;\n \n let i =0;\n for (let row of sample) {\n \n values.push(\n await extractor(embeddingAccessor(row), {\n pooling: \"mean\",\n normalize: true\n })\n );\n \n await bar.progress(i++, sample.length);\n // console.log(\"i\", i);\n }\n console.log(\"progress completed in\", performance.now() - before);\n bar.resolve(values);\n } else {\n // Not computing embeddgins, return nothing\n yield pleaseEnableComputeEmbeddings();\n }\n}","pinned":true,"hidden":false,"mode":"ojs"},{"id":659,"value":"## Appendix","pinned":false,"hidden":false,"mode":"md"},{"id":650,"value":"function pleaseEnableComputeEmbeddings() {\n return Object.assign(htl.html`Please enable compute embeddings`, {\n value: []\n });\n}","pinned":true,"hidden":false,"mode":"ojs"},{"id":418,"value":"sample = selected","pinned":true,"hidden":false,"mode":"ojs"},{"id":631,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":415,"value":"embeddings = {\n if (computeEmbeddings) {\n return embeddings_computed;\n } else {\n return embeddings_cached;\n }\n}","pinned":true,"hidden":false,"mode":"ojs"},{"id":407,"value":"embeddings_cached = FileAttachment(\"embeddings_v2.json\").json().then(res => {\n for (let row of res) \n row.data = new Float32Array(Array.from(Object.values(row.data)));\n return res;\n})\n","pinned":true,"hidden":false,"mode":"ojs"},{"id":169,"value":"embeddingAccessor = (d) => \n// [d.paper_type].join(\"\\n\\n\")\n computeEmbeddingsUsing.map(a => d[a]).join(\"\\n\\n\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":746,"value":"papersHighlighted = drSelection.drag\n ? dataToPlot.filter(\n (d) =>\n d.x >= drSelection.drag.x[0] &&\n d.x <= drSelection.drag.x[1] &&\n d.y >= drSelection.drag.y[0] &&\n d.y <= drSelection.drag.y[1]\n )\n : dataToPlot","pinned":true,"hidden":false,"mode":"ojs"},{"id":739,"value":"vegaScatterPlot = {\n let chart = vl\n .markCircle({ size: 100 })\n .encode(\n vl.x().fieldQ(\"x\").axis(null),\n vl.y().fieldQ(\"y\").axis(null),\n vl.color().fieldN(colorBy)\n )\n .width(Math.max(320, width - 150))\n .height(300)\n .data(dataToPlot);\n\n // console.log(\"vega spec interactive =\", interactive);\n\n if (interactive) {\n // console.log(\"vega interactive!\", interactive);\n const hover = vl\n .selectPoint(\"hover\")\n .nearest(true)\n .on(\"mouseover,pointerover,click,pointerup\")\n .init({ x: [], y: [] });\n const drag = vl\n .selectInterval(\"drag\")\n .init({ x: [], y: [] });\n\n chart = chart\n .params(hover, drag)\n .encode(\n vl.stroke().if(hover, vl.value(\"black\")).value(null),\n vl.size().if(vl.or(hover, drag), vl.value(100)).value(50),\n vl.tooltip([\n \"title\",\n \"paper_type_name\",\n \"event_title\",\n \"session_title\",\n \"abstract\"\n ])\n );\n }\n\n return chart;\n}","pinned":true,"hidden":false,"mode":"ojs"},{"id":106,"value":"reducerResult = Generators.observe(\n worker(\n function* dr({ method, data, parameters }) {\n yield* self.druid[method].generator(data, {}); // show the steps\n },\n {\n method: druid_method,\n data: embeddings.map((row) => row.data),\n parameters: reducerParams\n },\n `\nimportScripts(${JSON.stringify(\n await require.resolve(\"@saehrimnir/druidjs@^0.7.3\")\n )});\n`\n )\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":107,"value":"dataToPlot = sample.map((row, id) => ({\n ...row,\n x: reducerResult[id][0],\n y: reducerResult[id][1]\n}))","pinned":true,"hidden":false,"mode":"ojs"},{"id":115,"value":"$32","pinned":true,"hidden":false,"mode":"ojs"},{"id":1429,"value":"## Semantic Search","pinned":false,"hidden":false,"mode":"md"},{"id":1434,"value":"embeddingsHash = Object.fromEntries(papers.map((d, i) => [d.UID, embeddings[i]]))","pinned":true,"hidden":false,"mode":"ojs"},{"id":1298,"value":"viewof query = Inputs.text({label: \"Semantic search using embeddings\", value: \"\"})","pinned":false,"hidden":false,"mode":"ojs"},{"id":1327,"value":"Inputs.table(semanticSearchResults)","pinned":true,"hidden":false,"mode":"ojs"},{"id":1300,"value":"queryEmbedding = await extractor(query, {\n pooling: \"mean\",\n normalize: true\n })","pinned":true,"hidden":false,"mode":"ojs"},{"id":1302,"value":"semanticSearch = cosineSimilarity(queryEmbedding.data, sample.map(d => embeddingsHash[d.UID].data))","pinned":true,"hidden":false,"mode":"ojs"},{"id":1319,"value":"semanticSearchResults = semanticSearch\n .sort((a, b) => b.score - a.score)\n .map(({ i, score }) => ({\n score,\n title: sample[i].title,\n ...sample[i]\n }))","pinned":true,"hidden":false,"mode":"ojs"},{"id":1293,"value":"// https://github.com/xenova/transformers.js/blob/main/examples/semantic-image-search-client/src/app/worker.js#L46\nfunction cosineSimilarity(query_embeds, database_embeds) {\n const EMBED_DIM = query_embeds.length;\n const numDB = database_embeds.length;\n const similarityScores = new Array(numDB);\n\n for (let i = 0; i < numDB; ++i) {\n // const startOffset = i * EMBED_DIM;\n const dbVector = database_embeds[i];\n\n let dotProduct = 0;\n let normEmbeds = 0;\n let normDB = 0;\n\n for (let j = 0; j < EMBED_DIM; ++j) {\n const embedValue = query_embeds[j];\n const dbValue = dbVector[j];\n\n dotProduct += embedValue * dbValue;\n normEmbeds += embedValue * embedValue;\n normDB += dbValue * dbValue;\n }\n\n similarityScores[i] = {i, score: dotProduct / (Math.sqrt(normEmbeds) * Math.sqrt(normDB))};\n }\n\n return similarityScores;\n}\n","pinned":true,"hidden":false,"mode":"ojs"},{"id":979,"value":"druid = \"🤯\" // placeholder for Observable to allow the *dr* function to exist; the worker uses importScripts","pinned":true,"hidden":false,"mode":"ojs"},{"id":575,"value":"import {conditionalShow} from \"@john-guerra/conditional-show\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":734,"value":"import {vegaSelected} from \"@john-guerra/vega-selected\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":117,"value":"import { worker } from \"@fil/worker\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":118,"value":"import { form } from \"@mbostock/form-input\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":23,"value":"import {FacetedSearch} from \"@john-guerra/faceted-search\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":256,"value":"import {progress} from \"@mootari/displaying-progress\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":14,"value":"import {navio} from \"@john-guerra/navio\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":691,"value":"import {vl} from \"@john-guerra/vega-lite-api-v5\"","pinned":true,"hidden":false,"mode":"ojs"},{"id":58,"value":"transformers = import(\"https://cdn.jsdelivr.net/npm/@xenova/transformers@2.6.0\")","pinned":true,"hidden":false,"mode":"ojs"},{"id":52,"value":"### Transformers.js","pinned":false,"hidden":false,"mode":"md"},{"id":59,"value":"extractor = await transformers.pipeline(\n \"feature-extraction\",\n \"Xenova/all-MiniLM-L6-v2\",\n\n // \"allenai/specter2\",\n { quantized: false }\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":3,"value":"dicPapers = FileAttachment(\"paper_list_ieeevis_2023_from_github@1.json\")\n .json()\n // .then((res) =>\n // Array.from(Object.entries(res)).map(([UID, d]) => {\n // d.UID = UID;\n // d.time_stamp_parsed = new Date(d.time_stamp);\n // d.thumb = `https://ieeevis.b-cdn.net/vis_2023/paper_images_small/${d.UID}.png`;\n // return d;\n // })\n // )","pinned":true,"hidden":false,"mode":"ojs"},{"id":1641,"value":"papers.filter(d => d.abstract.toUpperCase().indexOf(\"EMBEDDINGS\") != -1)","pinned":true,"hidden":false,"mode":"ojs"},{"id":230,"value":"papers= FileAttachment(\"papers_vis2023@2.json\")\n .json()\n .then((res) =>\n res.map((d, i) => {\n d.time_stamp_parsed = new Date(d.time_stamp);\n d.thumb = `https://ieeevis.b-cdn.net/vis_2023/paper_images_small/${d.UID}.png`;\n d.url = `https://content.ieeevis.org/year/2023/paper_${d.UID}.html`\n\n // Override any attributes with the original values from github\n // This is for getting the actual abstract\n return Object.assign(d, dicPapers[d.UID]);\n })\n )","pinned":true,"hidden":false,"mode":"ojs"},{"id":361,"value":"Experiments computing the embeddings","pinned":false,"hidden":false,"mode":"md"},{"id":60,"value":"// embeddings2 = {\n// const before = performance.now();\n\n// const res = await Promise.all(\n// sample.map((row) =>\n// extractor(embeddingAccessor(row), {\n// pooling: \"mean\",\n// normalize: true\n// })\n// )\n// );\n\n// console.log(\"completed in\", performance.now() - before);\n// return res;\n// }","pinned":true,"hidden":false,"mode":"ojs"},{"id":285,"value":"// Generators.observe((notify) => {\n// const before = performance.now();\n// const res = [];\n// notify(res);\n// sample.map((row, i) => {\n// extractor(embeddingAccessor(row), {\n// pooling: \"mean\",\n// normalize: true\n// }).then((rowRes) => {\n// console.log(\"got result \", i)\n// res.push(rowRes);\n\n// if (res.length> sample.length-1) \n// console.log(\"Generator duration\", performance.now()-before);\n// notify(res);\n// });\n// });\n// })","pinned":true,"hidden":false,"mode":"ojs"},{"id":114,"value":"// {\n// const plt = Plot.plot({\n// marginTop: 50,\n// marginBottom: 50,\n// inset: 50,\n// width: width,\n// height: width,\n// marks: [\n// Plot.image(dataToPlot, {\n// x: \"x\",\n// y: \"y\",\n// src: (x) => x.thumb1,\n// width: 50,\n// title: \"x\"\n// }),\n// Plot.tip(\n// dataToPlot,\n// Plot.pointer({\n// stroke: \"paper_type\",\n\n// // channels: { Note: \"Note\", sport: \"sport\" },\n\n// x: \"x\",\n// y: \"y\",\n// title: embeddingAccessor\n// })\n// )\n// ]\n// });\n// const selection = d3.select(plt);\n// selection.on(\"mousemove\", (evt) => {\n// const tipdata = plt.value;\n// if (tipdata) {\n// const image = selection.select(`image[href='${tipdata.thumb1}']`);\n// image.raise();\n// }\n// });\n\n// return plt;\n// }","pinned":true,"hidden":false,"mode":"ojs"},{"id":730,"value":"$33","pinned":false,"hidden":false,"mode":"ojs"},{"id":1598,"value":"## Clusters","pinned":false,"hidden":false,"mode":"md"},{"id":1559,"value":"clusterEmbedding = require('@epivecs/cluster_embedding@0.1.11')","pinned":true,"hidden":false,"mode":"ojs"},{"id":1586,"value":"viewof maxClusters = Inputs.range([1, 20], {step: 1, label: \"Number of clusters\"})","pinned":true,"hidden":false,"mode":"ojs"},{"id":1560,"value":"viewof method = Inputs.select([\"kmc+pca\", \"kmc+sammon\", \"kmc+tsne\", \"kmc+umap\"], {label: \"Cluster Method\", value: \"kmc+sammon\"})\n","pinned":true,"hidden":false,"mode":"ojs"},{"id":1565,"value":"clusters = clusterEmbedding.clusterEmbed(\n embeddings.map((d) => d.data),\n maxClusters,\n method\n)","pinned":true,"hidden":false,"mode":"ojs"},{"id":1613,"value":"JSON.stringify(clusters.labels)","pinned":true,"hidden":false,"mode":"ojs"},{"id":1570,"value":"viewof cluster = Inputs.range([0, maxClusters], {step: 1})","pinned":true,"hidden":false,"mode":"ojs"},{"id":1583,"value":"Inputs.table(clusterPapers)","pinned":true,"hidden":false,"mode":"ojs"},{"id":1571,"value":"clusterPapers = {\n const indexes = clusters.labels.reduce((p, d, i) => {\n if (d === cluster) p.push(i);\n return p;\n }, []);\n \n return indexes.map((i) => papers[i]);\n}","pinned":true,"hidden":false,"mode":"ojs"}],"files":"$W34","stdlib":"1","titling":"auto"}} 2f:"$2e:body"