2f:Tede,dist_plot = { // calculate filtered data to plot based on user's ASes selection let data_to_plot = data_fetched_joined.filter(e=>e.xlabel == bias_dimension_labels_new[bias_dimensions_selection])[0]; const data_to_plot_filtered = {}; data_to_plot_filtered['feature'] = data_to_plot['feature']; data_to_plot_filtered['xlabel'] = data_to_plot['xlabel']; data_to_plot_filtered['ylabel'] = data_to_plot['ylabel']; if (data_to_plot['bars'] !== undefined) { data_to_plot_filtered['bars'] = {}; let filtered_test_data = Object.keys(data_to_plot['bars']).filter(e => selected_monitor_sets_mapped.includes(e)); for (let i=0; i selected_monitor_sets_mapped.includes(e)); for (let i=0; i { for (let i=0; i< Object.keys(colors).length; i++){ if (key.includes(Object.keys(colors)[i])) { selected_colors[Object.keys(colors)[i]] = colors[Object.keys(colors)[i]] } } }); if (data_to_plot_filtered["bars"] !== undefined) { return Plot.plot({ width: 1200, x: {axis: null}, y: {tickFormat: "s", domain: [0, 1]}, color: {domain: Object.keys(selected_colors), range: Object.values(selected_colors), legend: true}, marks: [ Plot.barY(dist_plot_data_formatted_filtered, { x: "AS", y: "value", fill: d => selected_colors[d["AS"]], fx: d => d[bias_dimensions_selection], // show value when hovered in bar plot title: d => d.value, sort: {x: null, fx: "y", order: "descending"},//, reduce: "max"} }), Plot.axisX({ tickFormat: d => d[bias_dimensions_selection], // tickRotate: -45, label: bias_dimensions_selection, }), Plot.ruleY([0]), ], }) } else { return Plot.plot({ width: 1200, marginBottom: 60, color: {domain: Object.keys(selected_colors), range: Object.values(selected_colors), legend: true}, x: { type: "log", tickRotate: -45, tickFormat: (d) => { const exponent = Math.floor(Math.log10(d)); const base = d / 10 ** exponent; return base.toFixed(1) + "e" + exponent; }, // tickFormat: (d, i, ticks) => { // const exponent = Math.floor(Math.log10(d)); // const base = d / 10 ** exponent; // const multiples = [0, 1, 2, 9, 10, 30, 200, 900, 1000, 2000, 9000, 10000, ]; // if (multiples.includes(i)) { // return base.toFixed(1) + "e" + exponent; // } // return null; // }, }, y: { grid: true, label: `↑ ${data_to_plot_filtered['ylabel']}` }, marks: [ Plot.ruleY([0]), Plot.lineY(dist_plot_data_formatted_filtered, {x: data_to_plot_filtered['xlabel'], y: data_to_plot_filtered['ylabel'], z: "AS", stroke: "AS", sort: {x: null, color: null, z: "AS"} }) ] }) } }30:T502,Now let's explain what we see in the table in more detail. The first column, as mentioned above, is the bias dimension each row corresponds to. Now every bias dimension, takes different values, either numerical or categorical. This is what the "Value" column represents. For the numerical values, data are divided into bins, with their ranges differ with respect to every dimension. For example, for the first dimension (number of ASNs), the first bin is 1-3, the second bin is 3-9, and it goes on until the last bin that is 16706-49213, because this particular features takes values from 1 to 49213. So, for every different bin of each feature, the third column, namely difference, shows the difference in the distributions between All ASes and each platform, for the particular dimension and each value that each row represents. Moving on to categorical values, the value column corresponds to the different values that each bias dimension can have, and the third column shows the difference in the distributions, as explained before, for that value. With the filter bar above the table, the user can search the table using keywords, for example by entering a bias dimension name, they will get the corresponding row(s) that shows difference for that dimension and its values.31:T935,// format data for table input data_monitor_set_formatted = { let dataArr = []; let max_; let hasNumber = /\d/; const data_fetched_table = (monitor_set_selection.includes("My")) ? ((Object.keys(data_fetched_monitor_set).length > 0) ? data_fetched_monitor_set["Custom list"] : 0 ) : data_fetched_monitor_set[monitors_index[monitor_set_selection]]; for (let i=0; i parseInt(elem, 10)); if (hasNumber.test(Object.keys(Object.values(data_fetched_table)[i])[j]) && Object.keys(Object.values(data_fetched_table)[i])[j].includes("-") && newObj['Bias Dimension']!= "AS hegemony" && newObj['Bias Dimension']!= "Country influence (CTI origin)"&& newObj['Bias Dimension']!= "Country influence (CTI top)"){ newObj['Value'] = Object.keys(Object.values(data_fetched_table)[i])[j].replace("-", " ").split(" ").map(elem=> parseInt(elem, 10)); newObj['Value'] = newObj['Value'].toString().replace(",", ", "); newObj['Value'] = "[" + newObj['Value'] + "]"; } else if ( newObj['Bias Dimension'] == "AS hegemony" || newObj['Bias Dimension']== "Country influence (CTI origin)" || newObj['Bias Dimension']== "Country influence (CTI top)") { newObj['Value'] = Object.keys(Object.values(data_fetched_table)[i])[j].split(/-(?=\d+\.)/).map(num => parseFloat(num).toExponential(2)); newObj['Value'] = newObj['Value'].toString().replace(",", ", "); newObj['Value'] = "[" + newObj['Value'] + "]"; } else if (hasNumber.test(Object.keys(Object.values(data_fetched_table)[i])[j]) && !Object.keys(Object.values(data_fetched_table)[i])[j].includes("-")){ newObj['Value'] = Object.keys(Object.values(data_fetched_table)[i])[j].split(" ").map(elem=> parseInt(elem, 10)); } else { newObj['Value'] = Object.keys(Object.values(data_fetched_table)[i])[j]; } newObj['Difference'] = parseFloat(Object.values(Object.values(data_fetched_table)[i])[j]); dataArr.push(newObj); } } return dataArr; } 32:T404,async function createData() { // const d = new Date(); // const t0 = d.getTime(); const ids = ["Atlas", "RIS", "RouteViews", "RIS&RouteViews"]; // Array of ids let bias = {}; const responses = await Promise.all( ids.map(async id => { bias[id] = await get_imp_bias(id); // Send request for each id }) ); let dataNew = []; for (let i=0; i { distr[id] = (id.startsWith("My"))? await get_dist_plots_custom(custom_sets_dict[id], set_numbers_type): await get_dist_plots_imps(id) ; // Send request for each id }) ); let dist_plots_fetched_data_custom = distr["My 1st Custom Set"] let dist_plots_fetched_data_custom1 = distr["My 2nd Custom Set"] let dist_plots_fetched_data_custom2 = distr["My 3rd Custom Set"] let dist_plots_fetched_data_all = distr["all"] let dist_plots_fetched_data_ATLAS = distr["Atlas"] let dist_plots_fetched_data_RIS = distr["RIS"] let dist_plots_fetched_data_RouteViews = distr["RouteViews"] let dist_plots_fetched_data_RISRouteViews = distr["RIS&RouteViews"] for (let i=0; ie.xlabel == bias_dimension_labels_new[bias_dimensions_selection])[0]; let dataArr = []; if (data_to_plot["bars"] !== undefined) { let data_to_plot_bars = data_to_plot['bars']; for (let j=0; j e === \"Country\"),1);\n biasArr.splice(biasArr.findIndex(e => e === \"ASDB C1L2\"),1);\n const select = Inputs.select(biasArr, {label: \"Bias dimensions\", value: biasArr});\n\n select.style.width = \"100%\";\n return select;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":161,"value":"$2f","pinned":false,"hidden":false,"mode":"ojs"},{"id":138,"value":"For example, while examining the table for RIPE RIS platform, in the \"Network type\" dimension, we spot a small difference (0.02%) in the value of \"Network Services\", which means that the government networks are equally represented in both All ASes and the platform. \nOn the contrary, in the \"number of ASNs\" dimension, we spot a big difference (-60%) for values from 1 to 3 (#ASNs), which means that All ASes in general have really less networks with number of asns among 1-3, than the platform. So, that particular values in this dimension are over-represented by RIPE RIS. \n","pinned":false,"hidden":false,"mode":"md"},{"id":6,"value":"viewof monitor_set_selection = {\n const select = Inputs.select(monitor_sets, {label: \"Monitor Set\", value: monitor_sets});\n\n select.style.width = \"100%\";\n return select;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":17,"value":"viewof searched_data = {\n const search_field = Inputs.search(data_monitor_set_formatted, {label: 'Filter by: Bias Dimension, Value or Difference'});\n\n search_field.style.width = \"100%\";\n return search_field;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":27,"value":"viewof table_monitor_set = {\n return Inputs.table(searched_data, {\n \n // sort: \"TOTAL bias (selected)\",\n \n columns: [\n \"Bias Dimension\",\n \"Value\",\n \"Difference\"\n ],\n align: {\n \"Bias Dimension\": \"center\",\n \"Value\": \"center\",\n \"Difference\": \"center\"\n },\n header: {\n \"Bias Dimension\": \"Bias Dimension\",\n \"Value\": \"Value Range\",\n \"Difference\": \"Difference in %\"\n },\n\n format: {\n // \"Bias Dimension\": d => d['Bias Dimension'],\n // \"Value\": d => Object.keys(d),\n \"Difference\": d => d,\n },\n})\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":58,"value":"$30","pinned":false,"hidden":false,"mode":"md"},{"id":70,"value":"The importance of the differences in the distributions of the bias dimensions, for each one of their values, is in their usage as recommedations for platform owners and operators. In more detail, by seeing a platform monitors set or a custom monitors set having huge differences in certain values for one or more bias dimensions, the users should know that they cannot base their analyses or conclusions of various measurements on those features, as they might be biased, so their findings will be as well.\n\nFurthermore, those insights that are obtained from the table, can be used as recommendations for choosing another set of monitors, in order to have more complete view of all the networks, and in result a more uniform distribution chosen from the whole set of ASNs. \nAlso, for the platform owners, the difference table could be a useful indicator of how and what new monitors to add to their platforms, having as purpose to mitigate this difference, and hence the bias that it creates.","pinned":false,"hidden":false,"mode":"md"},{"id":218,"value":"Let's shed light into some example distribution differences, taking into consideration the smallest and largest positive, and then negative difference. \n\nBeginning with the positive differences, that mean the platform is underrepresented for the particular row of the bias cause table, the min and max values are presented below: ","pinned":false,"hidden":false,"mode":"md"},{"id":256,"value":"smallest_pos_diff_widget = {\n if (only_pos.length !== 0) {\n return html`\n

${Math.min(...only_pos)} is the min positive difference for ${data_monitor_set_formatted.map(el => el['Difference'] == Math.min(...only_pos) ? el['Bias Dimension'] : '')} with value: ${data_monitor_set_formatted.map(el => el['Difference'] == Math.min(...only_pos) ? el['Value'] : '')}

\n `\n } else {\n return html`\n

No results

\n `\n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":300,"value":"nettype2 = FileAttachment(\"nettype@2.png\").image()","pinned":false,"hidden":false,"mode":"ojs"},{"id":310,"value":"In this case, the platform of RIPE RIS is represented well enough, regarding ASes that are of Network type \"Network Service\". Although in the same plot, we can spot how much overrepresented the platform is in \"NSP\" Network type, and underrepresented in the \"Not Disclosed\" Network type, respectively.","pinned":false,"hidden":false,"mode":"md"},{"id":222,"value":"bigg_pos_diff_widget = {\n if (only_pos.length !== 0) {\n return html`\n

${Math.max(...only_pos)} is the max positive difference for ${data_monitor_set_formatted.map(el => el['Difference'] == Math.max(...only_pos) ? el['Bias Dimension'] : '')} with value: ${data_monitor_set_formatted.map(el => el['Difference'] == Math.max(...only_pos) ? el['Value'] : '')}

\n `\n } else {\n return html`\n

No results

\n `\n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":304,"value":"traf1 = FileAttachment(\"traf@1.png\").image()","pinned":false,"hidden":false,"mode":"ojs"},{"id":313,"value":"In the plot above, we spot the max positive distribution difference in the Traffic ratio dimension, regarding the ASes of type: \"Not Disclosed\". We can see that the RIS platform is underrepresented regarding such ASes, in contrast to \"Heavy Outbound\", where the distributions are very similar.","pinned":false,"hidden":false,"mode":"md"},{"id":290,"value":"Moving on to the negative differences, that mean the platform is overrepresented, the min and max values are presented below: ","pinned":false,"hidden":false,"mode":"md"},{"id":283,"value":"smallest_neg_diff_widget = {\n if (only_negs.length !== 0) {\n return html`\n

${Math.min(...only_negs)} is the min negative difference for ${data_monitor_set_formatted.map(el => el['Difference'] == Math.min(...only_negs) ? el['Bias Dimension'] : '')} with value: ${data_monitor_set_formatted.map(el => el['Difference'] == Math.min(...only_negs) ? el['Value'] : '')}

\n `\n } else {\n return html`\n

No results

\n `\n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":306,"value":"peers = FileAttachment(\"peers.png\").image()","pinned":false,"hidden":false,"mode":"ojs"},{"id":315,"value":"The plot above is a CDF plot, so it differs from the bar plots we have seen so far. Here, we study the difference in distributions in bins, and in this case we spot the largest one in the first bin, which is from 1 to 3, regarding number of peers dimension.","pinned":false,"hidden":false,"mode":"md"},{"id":246,"value":"bigg_neg_diff_widget = {\n if (only_negs.length !== 0) {\n return html`\n

${Math.max(...only_negs)} is the max negative difference for ${data_monitor_set_formatted.map(el => el['Difference'] == Math.max(...only_negs) ? el['Bias Dimension'] : '')} with value: ${data_monitor_set_formatted.map(el => el['Difference'] == Math.max(...only_negs) ? el['Value'] : '')}

\n `\n } else {\n return html`\n

No results

\n `\n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":365,"value":"ashege1 = FileAttachment(\"ashege@1.png\").image()","pinned":false,"hidden":false,"mode":"ojs"},{"id":319,"value":"Finally, the last plot shows another CDF plot, for the bias dimension of AS hegemony, which takes small values in general. The least negative difference, is in the bin [4.03e-4, 3.18e-3]. ","pinned":false,"hidden":false,"mode":"md"},{"id":321,"value":"To conclude, the bias causes can shed light into the reasons we spot bias, in our bias analysis on the monitoring platforms. For example, by looking down at a dimension which has a large bias value in the radar plot, we can take a close look in the bar plot or CDF plot of that dimension and understand that we have big differences in the distributions of the most values of that dimension. For example, take a look in the radar plot below:","pinned":false,"hidden":false,"mode":"md"},{"id":325,"value":"radarplot = FileAttachment(\"radarplot.PNG\").image()","pinned":false,"hidden":false,"mode":"ojs"},{"id":326,"value":"The biggest bias score is in total number of neighbors, and using the interactive plot in the top of the article, by choosing this feature we can see that there are differences in the distributions of the CDF plot. Another example, regarding a bar plot which may be more easy to understand, is the Traffic ratio dimension, where we can see the bars differ in the most of the values.\n\nSo, we believe it is essential, except from quantifying and analyzing the bias, to understand in more detail where this bias comes from, in order to avoid it or even take measures trying to mitigate it.","pinned":false,"hidden":false,"mode":"md"},{"id":41,"value":"## CODE","pinned":false,"hidden":false,"mode":"md"},{"id":63,"value":"// top3 = table_monitor_set.slice().sort(function (a, b) { return b.difference - a.difference; }).slice(0, 3);\n// st = table_monitor_set.sort((x,y) => y.difference-x.difference);\ntable_monitor_set","pinned":false,"hidden":false,"mode":"ojs"},{"id":220,"value":"diff_only_arr = data_monitor_set_formatted.map(el => el['Difference']);","pinned":false,"hidden":false,"mode":"ojs"},{"id":265,"value":"function getPositiveNumbers(array) {\n return array.filter(value => value > 0);\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":259,"value":"only_pos = getPositiveNumbers(diff_only_arr);\n","pinned":false,"hidden":false,"mode":"ojs"},{"id":234,"value":"function getNegativeNumbers(array) {\n return array.filter(value => value < 0);\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":239,"value":"only_negs = getNegativeNumbers(diff_only_arr)","pinned":false,"hidden":false,"mode":"ojs"},{"id":14,"value":"viewof checkboxes_monitor_sets = {\n const checkboxes_monitor_sets = Inputs.checkbox(monitor_sets, {label: \"Monitor sets\", value: monitor_sets});\n checkboxes_monitor_sets.style.width = \"100%\";\n return checkboxes_monitor_sets;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":12,"value":"selected_monitor_sets = monitor_sets.filter(e => checkboxes_monitor_sets.includes(e))","pinned":false,"hidden":false,"mode":"ojs"},{"id":4,"value":"monitors_index = Object({\n \"RIPE RIS\": \"RIS\",\n \"RIPE Atlas\": \"Atlas\",\n \"RouteViews\": \"RouteViews\"})","pinned":false,"hidden":false,"mode":"ojs"},{"id":10,"value":"monitor_sets = [\"RIPE RIS\"]","pinned":false,"hidden":false,"mode":"ojs"},{"id":25,"value":"viewof set_numbers_type = {\n const types_array = [\"ASNs\", \"probe IDs\"];\n const set_numbers_type = Inputs.radio(types_array, {label: \"Select the type of list of numbers\", value: types_array[0]});\n set_numbers_type.style.width = \"100%\";\n return set_numbers_type;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":19,"value":"$31","pinned":false,"hidden":false,"mode":"ojs"},{"id":21,"value":"// fetch initial data from endpoint\ndata_fetched_monitor_set = {\n const monitor = monitors_index[monitor_set_selection];\n let mtype = (set_numbers_type === \"ASNs\") ? \"asn\" : \"probe\";\n let arr = [];\n if (monitor == 'Atlas' || monitor == 'RIS' || monitor == 'RouteViews') {\n arr = fetch(\"https://ai4netmon.csd.auth.gr/api/bias/cause/\"+monitors_index[monitor_set_selection]).then((response) => \n response.json());}\n return arr;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":146,"value":"bias_dimension_labels_to_show = ({\n \"RIR region\": \"RIR region\",\n \"Country\": \"Country\",\n \"Continent\": \"Continent\",\n \"Customer cone (#ASNs)\": \"Customer cone (#ASNs)\",\n \"Customer cone (#pfxs)\": \"Customer cone (#pfxs)\",\n \"Customer cone (#addresses)\": \"Customer cone (#addresses)\",\n \"AS Hegemony\": \"AS Hegemony\",\n \"CTI origin\": \"CTI origin\",\n \"CTI top\": \"CTI top\",\n \"#neighbors (total)\": \"#neighbors (total)\",\n \"#peers\": \"#peers\",\n \"#customers\": \"#customers\",\n \"#providers\": \"#providers\",\n \"#IXPs\": \"#IXPs\",\n \"#facilities\": \"#facilities\",\n \"Peering policy\": \"Peering policy\",\n \"ASDB C1L1\": \"ASDB C1L1\",\n \"ASDB C1L2\": \"ASDB C1L2\",\n \"Net. type\": \"Net. type\",\n \"Traffic ratio\": \"Traffic ratio\",\n \"Traffic volume\": \"Traffic volume\",\n \"Scope\": \"Scope\",\n \"Personal ASN\": \"Personal ASN\"\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":151,"value":"bias_dimension_labels_new = ({\n \"RIR region\": \"RIR region\",\n \"Country\": \"Location (country)\",\n \"Continent\": \"Location (continent)\",\n \"Customer cone (#ASNs)\": \"Customer cone (#ASNs)\",\n \"Customer cone (#pfxs)\": \"Customer cone (#prefixes)\",\n \"Customer cone (#addresses)\": \"Customer cone (#addresses)\",\n \"AS Hegemony\": \"AS hegemony\",\n \"CTI origin\": \"Country influence (CTI origin)\",\n \"CTI top\": \"Country influence (CTI top)\",\n \"#neighbors (total)\": \"#neighbors (total)\",\n \"#peers\": \"#neighbors (peers)\",\n \"#customers\": \"#neighbors (customers)\",\n \"#providers\": \"#neighbors (providers)\",\n \"#IXPs\": \"#IXPs (PeeringDB)\",\n \"#facilities\": \"#facilities (PeeringDB)\",\n \"Peering policy\": \"Peering policy (PeeringDB)\",\n \"ASDB C1L1\": \"ASDB C1L1\",\n \"ASDB C1L2\": \"ASDB C1L2\",\n \"Net. type\": \"Network type (PeeringDB)\",\n \"Traffic ratio\": \"Traffic ratio (PeeringDB)\",\n \"Traffic volume\": \"Traffic volume (PeeringDB)\",\n \"Scope\": \"Scope (PeeringDB)\",\n \"Personal ASN\": \"Personal ASN\"\n})","pinned":false,"hidden":false,"mode":"ojs"},{"id":153,"value":"function get_imp_bias(imp) {\n return fetch(\"https://ai4netmon.csd.auth.gr/api/bias/\"+imp).then((response) => response.json())\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":155,"value":"data_fetched = createData()","pinned":false,"hidden":false,"mode":"ojs"},{"id":148,"value":"$32","pinned":false,"hidden":false,"mode":"ojs"},{"id":157,"value":"viewof checkboxes_bias_dimensions = {\n const checkboxes_bias_dimensions = Inputs.checkbox(Object.keys(bias_dimension_labels_new), {label: \"Bias dimensions\", value: Object.keys(bias_dimension_labels_new)});\n checkboxes_bias_dimensions.style.width = \"100%\";\n return checkboxes_bias_dimensions;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":144,"value":"bias_dimensions = data_fetched.map(o=>o.tag).filter(e => checkboxes_bias_dimensions.includes(bias_dimension_labels_to_show[e]))","pinned":false,"hidden":false,"mode":"ojs"},{"id":167,"value":"monitor_sets_Obj = Object({\n \"RIPE Atlas\": \"Atlas\",\n \"RIPE RIS\": \"RIS\",\n \"RouteViews\": \"RouteViews\",\n \"RIS&RouteViews\": \"RIS&RouteViews\"\n })","pinned":false,"hidden":false,"mode":"ojs"},{"id":165,"value":"selected_monitor_sets_mapped = {\n const newArr = [];\n newArr.push(\"All ASes\");\n for (let i=0; i response.json());\n return data[0]\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":181,"value":"async function get_dist_plots_custom(arg_custom_set, arg_set_numbers_type){\n if (arg_custom_set.length>0){\n let mtype = (arg_set_numbers_type === \"ASNs\")? \"asn\":\"probe\"\n // console.log(mtype)\n let data = await fetch(\"https://ai4netmon.csd.auth.gr/api/distributions/custom/\"+mtype+\"/?\"+mtype+\"=\"+arg_custom_set.replaceAll(\",\",\"&\"+mtype+\"=\")).then((response) => response.json());\n // console.log(data[0])\n return data[0];\n } else {\n return null; \n }\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":183,"value":"viewof custom_set = {\n const custom_set = Inputs.text({label: `Custom Set #1 (${set_numbers_type})`});\n custom_set.style.width = \"100%\";\n return custom_set;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":185,"value":"viewof custom_set1 = {\n const custom_set1 = Inputs.text({label: `Custom Set #2 (${set_numbers_type})`});\n custom_set1.style.width = \"100%\";\n return custom_set1;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":187,"value":"viewof custom_set2 = {\n const custom_set2 = Inputs.text({label: `Custom Set #3 (${set_numbers_type})`});\n custom_set2.style.width = \"100%\";\n return custom_set2;\n}","pinned":false,"hidden":false,"mode":"ojs"},{"id":172,"value":"$33","pinned":false,"hidden":false,"mode":"ojs"},{"id":170,"value":"$34","pinned":false,"hidden":false,"mode":"ojs"},{"id":163,"value":"dist_plot_data_formatted_filtered = dist_plot_data_formatted.filter(e => selected_monitor_sets_mapped.includes(e['AS']))","pinned":false,"hidden":false,"mode":"ojs"}],"files":"$W35","stdlib":"1","titling":"auto"}