Fountaincoder/stuart.yaml

## stuart.yaml
name: Basic API call (TypeScript)
description: Performs a basic Excel API call using TypeScript.
host: EXCEL
api_set: {}
script:
  content: |
    $("#run").click(() => tryCatch(run));

    console.log("Ready");

    async function run() {
      const days = Number($("#days").val());
      const no_sims = Number($("#sims").val());
      const gencheck = $("#gencheck").is(":checked");
      const scale = Number($("#genscale").val()) / 100.0;
      const weekend_act = Number($("#weekend").val());
      const weekend = $("#wkdcheck").is(":checked");

      let schedule = [];

      if ($("#ccg").is(":checked")) {
        const ccgno = Number($("#ccgno").val());
        const ccgmin = Number($("#ccgmin").val());
        const ccgmax = Number($("#ccgmax").val());
        const ccglen = Number($("#ccglen").val());
        const ccgres = Number($("#ccgres").val());

        const e2 = Array.from(Array(ccgno).keys()).map((_) => ({
          func: spike,
          peak: () => sim_rand(ccgmin, ccgmax),
          length: () => sim_rand(1, ccglen),
          net: () => sim_rand(1, ccgres),
          when: () => sim_rand(1, days)
        }));
        schedule = schedule.concat(e2);
      }
      if ($("#ccg2").is(":checked")) {
        const ccg2no = Number($("#ccg2no").val());
        const ccg2txt = Number($("#ccg2txt").val());
        const ccg2conv = Number($("#ccg2conv").val());
        const ccg2len = Number($("#ccg2len").val());
        const ccg2res = Number($("#ccg2res").val());

        const e2 = Array.from(Array(ccg2no).keys()).map((_) => ({
          func: spike,
          peak: () => (ccg2conv / 100) * ccg2txt,
          length: () => sim_rand(2, ccg2len),
          net: () => ccg2res,
          when: () => sim_rand(1, days)
        }));
        schedule = schedule.concat(e2);
      }

      if ($("#covid").is(":checked")) {
        const cwhen = Number($("#cwhen").val());
        const cmin = Number($("#cmin").val());
        const cmax = Number($("#cmax").val());
        const cres = Number($("#cres").val());

        const e2 = [
          {
            func: spike,
            peak: () => sim_rand(cmin, cmax),
            length: () => 30,
            net: () => cres,
            when: () => cwhen
          }
        ];
        schedule = schedule.concat(e2);
      }

      if ($("#covid2").is(":checked")) {
        const cwhen = Number($("#c2when").val());
        const cmin = Number($("#c2min").val());
        const cmax = Number($("#c2max").val());
        const cres = Number($("#c2res").val());

        const e2 = [
          {
            func: spike,
            peak: () => sim_rand(cmin, cmax),
            length: () => 30,
            net: () => cres,
            when: () => cwhen
          }
        ];
        schedule = schedule.concat(e2);
      }

      if ($("#flu").is(":checked")) {
        const when = Number($("#wwhen").val());
        const len = Number($("#wlen").val());
        const extra = Number($("#wextra").val());
        const e2 = [
          {
            func: rise,
            days: () => len,
            net: () => extra,
            when: () => when
          }
        ];
        schedule = schedule.concat(e2);
      }

      if ($("#bckgnd").is(":checked")) {
        const bfrom = Number($("#bfrm").val());
        const bperc = Number($("#bperc").val());
        const e2 = [
          {
            func: trend,
            c: () => bfrom,
            m: () => (bperc * (bfrom + 1)) / (100 * days),
            when: () => 0
          }
        ];
        schedule = schedule.concat(e2);
      }

      if ($("#partners").is(":checked")) {
        const pnums = Number($("#pnums").val());
        const pscale = Number($("#pscale").val());
        const pperc = Number($("#pperc").val());

        const e2 = Array.from(Array(pnums).keys()).map(function(_) {
          let pfrom = pscale * 150;

          return {
            func: trend,
            c: () => pfrom,
            m: () => (pperc * (pfrom + 1)) / (100 * days),
            when: () => sim_rand(1, days)
          };
        });

        schedule = schedule.concat(e2);
      }

      console.log(schedule.length + " events scheduled");
      Excel.run(function(context) {
        console.log("Starting Sims");
        let sims = [];

        for (let i = 0; i < no_sims; i++) {
          const s = run_sim(schedule, weekend, weekend_act, scale, gencheck, days);
          const sync_s = s.arraySync();
          sims.push(sync_s);
        }
        console.log("Finished Sims");
        console.log("Calculate Statistics");

        const s2 = tf.tensor2d(sims);
        const s3 = tf.moments(s2, 0);
        const mean_s = s3["mean"].arraySync().map((x) => [x]);
        const cummean_s = s3["mean"]
          .cumsum()
          .arraySync()
          .map((x) => [x]);
        const stddev_s = s3["variance"]
          .sqrt()
          .arraySync()
          .map((x) => [x]);
        console.log("Updating working sheet");
        var sheet = context.workbook.worksheets.getItem("Simulations");

        var simrange = sheet.getRange("A1:A" + String(days + 1));
        simrange.values = [["Submissions: Sample Sim"]].concat(sims[0].map((x) => [x]));
        simrange.format.autofitColumns();

        var mrange = sheet.getRange("B1:B" + String(days + 1));
        mrange.values = [["Submissions: Mean of Sims"]].concat(mean_s);
        mrange.format.autofitColumns();

        var stdrange = sheet.getRange("C1:C" + String(days + 1));
        stdrange.values = [["Stddev"]].concat(stddev_s);
        stdrange.format.autofitColumns();

        var cumurange = sheet.getRange("D1:D" + String(days + 1));
        cumurange.values = [["Cumulative Submissions"]].concat(cummean_s);
        cumurange.format.autofitColumns();

        var chart1 = sheet.charts.add("Line", simrange, "auto");
        chart1.title.text = "Single Simulation: Submissions";

        var chart2 = sheet.charts.add("Line", mrange, "auto");
        chart2.title.text = "Mean of Simulations: Submissions";

        var chart3 = sheet.charts.add("Line", cumurange, "auto");
        chart3.title.text = "Mean of Simulations: Cumulative Submissions";

        chart1.setPosition("E1", "J16");
        chart2.setPosition("E18", "J32");
        chart3.setPosition("E34", "J48");
        console.log("finished");
        //chart.legend.position = "right"
        //chart.legend.format.fill.setSolidColor("white");
        //chart.dataLabels.format.font.size = 5;
        //chart.dataLabels.format.font.color = "black";

        return context.sync();
      });
    }
    function rise(ev, no_days = 365) {
      const days = ev["days"]();
      const net = ev["net"]();
      const when = ev["when"]();

      const y = tf.fill([days], net);
      return y.pad([[when, no_days - (when + days)]]);
    }

    function weekend_adjustment(no_days = 365, weekend) {
      //}, minScale = 0.6, maxScale = 0.7) {

      const minScale = weekend / 100 - 0.1;
      const maxScale = weekend / 100 + 0.1;
      const weeks = Math.floor(no_days / 7);
      const weekdays = tf.ones([weeks, 5]);
      const weekends = tf.randomUniform([weeks, 2], minScale, maxScale);
      const all_days = tf.concat([weekdays, weekends], (axis = 1)).flatten();
      const no_act_days = all_days.shape[0];
      return all_days.pad([[0, no_days - no_act_days]], 1);
    }

    function spike(ev, no_days = 365) {
      const peak = ev["peak"]();
      const len = ev["length"]();
      const net = ev["net"]();
      const when = ev["when"]();

      const l0 = (Math.log(peak) + 1) / len;
      const x = tf.range(0, len);
      const spike_out = tf.scalar(peak - net).mul(tf.exp(tf.scalar(-l0).mul(x)));
      const ele_spike_out = spike_out.add(tf.scalar(net));
      const start_padded_spike = ele_spike_out.pad([[when, 0]]);
      return start_padded_spike.pad([[0, no_days - (when + len)]], net);
    }

    function trend(ev, no_days = 365) {
      const m = ev["m"]();
      const c = ev["c"]();
      const when = ev["when"]();

      const x = tf.range(0, no_days);
      const trend_out = tf
        .scalar(m)
        .mul(x)
        .add(tf.scalar(c));
      return position(trend_out, when, no_days);
    }

    function position(data, when, no_days) {
      return data.pad([[when, no_days - (data.shape[0] + when)]]);
    }

    function run_sim(evts, wkdcheck, weekends, scale, gencheck, no_days = 365) {
      let yrs = tf.zeros([no_days]);
      for (var e in evts) {
        const event_days = evts[e]["func"](evts[e], no_days);
        yrs = yrs.add(event_days);
      }
      if (wkdcheck) {
        yrs = yrs.mul(weekend_adjustment(no_days, weekends));
      }
      if (gencheck) {
        yrs = tf.scalar(scale).mul(yrs);
      }
      return yrs;
    }

    function sim_rand(mn, mx) {
      const shell_out = (x) => x.arraySync()[0];
      return Math.floor(shell_out(tf.randomUniform([1], mn, mx)));
    }

    /** Default helper for invoking an action and handling errors. */
    async function tryCatch(callback) {
      try {
        await callback();
      } catch (error) {
        // Note: In a production add-in, you'd want to notify the user through your add-in's UI.
        console.error(error);
      }
    }
  language: typescript
template:
  content: "<section class=\"samples ms-font-m\">\n\t<h1>NHS-Login Forecasting</h1>\n\t<p> Please fill in the details of the simulation below (it scrolls, and has 7 Scenario types) and press the\n\t\t<b>Generate Forecast</b> button.</p>\n\t<label> No. Simulations to aggregate: </label> <input id=\"sims\" type=\"text\" size=\"3\" value=\"10\"><br><br>\n\t<button id=\"run\" class=\"ms-Button\"> <span class=\"ms-Button-label\">Generate Forecast</span></button>\n\n\t<h2> Settings </h2>\n\n\t<h3> General </h3>\n\t<label> Length of forecast: </label>\n\t<input id=\"days\" type=\"text\" size=\"3\" value=\"365\"><label><em> (days)</em></label><br>\n\t<input type=\"checkbox\" id=\"wkdcheck\"> <label> Weekend as % of weekday:  </label>\n\t<input id=\"weekend\" type=\"text\" size=\"3\" value=\"60\"><em> (1-100)+-10%</em><br>\n\t<input type=\"checkbox\" id=\"gencheck\"> <label> general scaling:  </label>\n\t<input id=\"genscale\" type=\"text\" size=\"3\" value=\"100\"><em> (1-500)</em><br>\n\t<p> <em>General scaling </em> is a multiplying factor for the entire forecast. It's a way of upgrading/downgrading\n\t\tthe values to reflect things such as a more postive view of NHS-app numbers. </p>\n\n\t<h3> <label>1. Spike Events  </label> <input type=\"checkbox\" id=\"ccg\"> </h3>\n\t\t<p><em> Spikes, randomly throughout period</em></p>\n\t\t<label> Number: </label> <input id=\"ccgno\" type=\"text\" size=\"3\" value=\"10\"><em> (events)</em><br>\n\t\t<label> Min Extra Submissions: </label>\n\t\t<input id=\"ccgmin\" type=\"text\" size=\"5\" value=\"4000\"><em> (persons)</em><br>\n\t\t<label> Max Extra Submissions: </label>\n\t\t<input id=\"ccgmax\" type=\"text\" size=\"5\" value=\"6000\"><em> (persons)</em><br>\n\t\t<label> Length: </label> <input id=\"ccglen\" type=\"text\" size=\"3\" value=\"2\"><em> (days)</em><br>\n\t\t<label> Residual after event up to: </label> <input id=\"ccgres\" type=\"text\" size=\"5\" value=\"500\">\n\t\t<em> (persons)</em>\n\n\t\t<h3> <label>2. CCG Events  </label> <input type=\"checkbox\" id=\"ccg2\"> </h3>\n\t\t\t<p><em> Notts-style CCG event spikes, randomly throughout period</em></p>\n\t\t\t<label> Number: </label> <input id=\"ccg2no\" type=\"text\" size=\"3\" value=\"10\"><em> (events)</em><br>\n\t\t\t<label> Txts: </label>\n\t\t\t<input id=\"ccg2txt\" type=\"text\" size=\"6\" value=\"100000\"><em> (persons)</em><br>\n\t\t\t<label> % Conversion: </label>\n\t\t\t<input id=\"ccg2conv\" type=\"text\" size=\"3\" value=\"6\"><em> (1-100)</em><br>\n\t\t\t<label> Length: </label> <input id=\"ccg2len\" type=\"text\" size=\"3\" value=\"2\"><em> (days)</em><br>\n\t\t\t<label> Residual after event up to: </label> <input id=\"ccg2res\" type=\"text\" size=\"5\" value=\"500\">\n\t\t\t<em> (persons)</em>\n\n\t\t\t<h3> <label>3. Partners  </label> <input type=\"checkbox\" id=\"partners\"> </h3>\n\t\t\t\t<p><em>How partners will grow over period. Assumption is linear growth. Co-op is \n\t\t\t\t\tone unit and equates to 150 submissions per day. We assume that these partners will have\n\t\t\t\t\ta significant impact as per co-op</em> </p>\n\t\t\t\t<label>Number of New significant partners in year: </label>\n\t\t\t\t<input id=\"pnums\" type=\"text\" size=\"3\" value=\"20\"><em> (companies)</em><br>\n\t\t\t\t<label>Max Company size when compared to the Co-op: </label>\n\t\t\t\t<input id=\"pscale\" type=\"text\" size=\"3\" value=\"1.0\"><em> </em><br>\n\t\t\t\t<label>% growth over year: </label> <input id=\"pperc\" type=\"text\" size=\"3\" value=\"10\"><em> (0-100)</em>\n\n\n\t\t\t\t<h3> <label>4. Background Growth  </label> <input type=\"checkbox\" id=\"bckgnd\"> </h3>\n\t\t\t\t\t<p><em>General background growth over period. Assumption is linear growth. </em></p>\n\t\t\t\t\t<label>Starting value: </label>\n\t\t\t\t\t<input id=\"bfrm\" type=\"text\" size=\"5\" value=\"10000\"><em> (persons)</em><br>\n\t\t\t\t\t<label>% growth over year: </label>\n\t\t\t\t\t<input id=\"bperc\" type=\"text\" size=\"3\" value=\"10\"><em> (0-100)</em><br>\n\n\n\t\t\t\t\t<h3> <label>5. Covid Lockdown</label> <input type=\"checkbox\" id =\"covid\"> </h3>\n\t\t\t\t\t\t<p><em>A 1st major spikes at a period in the future</em></p>\n\t\t\t\t\t\t<label> When:</label> <input id=\"cwhen\" type=\"text\" size=\"3\" value=\"0\"><em> (day num.)</em><br>\n\t\t\t\t\t\t<label> Min Extra Submissions: </label>\n\t\t\t\t\t\t<input id=\"cmin\" type=\"text\" size=\"5\" value=\"18000\"><em> (persons)</em><br>\n\t\t\t\t\t\t<label> Max Extra Submissions:</label>\n\t\t\t\t\t\t<input id=\"cmax\" type=\"text\" size=\"5\" value=\"20000\"><em> (persons)</em><br>\n\t\t\t\t\t\t<label> Residual after event:</label>\n\t\t\t\t\t\t<input id=\"cres\" type=\"text\" size=\"5\" value=\"7500\"><em> (persons)</em><br>\n\n\t\t\t\t\t\t<h3> <label>6. Covid 2nd Lockdown</label> <input type=\"checkbox\" id =\"covid2\"> </h3>\n\t\t\t\t\t\t\t<p><em>A 2nd major spikes at a period in the future</em></p>\n\t\t\t\t\t\t\t<label> When:</label>\n\t\t\t\t\t\t\t<input id=\"c2when\" type=\"text\" size=\"3\" value=\"100\"><em> (day num.)</em><br>\n\t\t\t\t\t\t\t<label> Min Extra Submissions: </label>\n\t\t\t\t\t\t\t<input id=\"c2min\" type=\"text\" size=\"5\" value=\"18000\"><em> (persons)</em><br>\n\t\t\t\t\t\t\t<label> Max Extra Submissions:</label>\n\t\t\t\t\t\t\t<input id=\"c2max\" type=\"text\" size=\"5\" value=\"20000\"><em> (persons)</em><br>\n\t\t\t\t\t\t\t<label> Residual after event:</label>\n\t\t\t\t\t\t\t<input id=\"c2res\" type=\"text\" size=\"5\" value=\"7500\"><em> (persons)</em><br>\n\n\t\t\t\t\t\t\t<h3> <label>7. Winter Rise</label> <input type=\"checkbox\" id=\"flu\"> </h3>\n\t\t\t\t\t\t\t\t<p> <em>A general rise in usage typically associated with a november normal flu epidemic</em>\n\t\t\t\t\t\t\t\t</p>\n\t\t\t\t\t\t\t\t<label> When: </label>\n\t\t\t\t\t\t\t\t<input id=\"wwhen\" type=\"text\" size=\"3\" value=\"100\"><label><em> (day num.)</em></label><br>\n\t\t\t\t\t\t\t\t<label> For how long: </label>\n\t\t\t\t\t\t\t\t<input id=\"wlen\" type=\"text\" size=\"3\" value=\"50\"><em> (days)</em><br>\n\t\t\t\t\t\t\t\t<label> Extra Submissions: </label>\n\t\t\t\t\t\t\t\t<input id=\"wextra\" type=\"text\" size=\"5\" value=\"1000\"><em> (persons)</em><br><br>\n\n\n</section>"
  language: html
style:
  content: |-
    section.samples {
        margin-top: 20px;
    }

    section.samples .ms-Button, section.setup .ms-Button {
        display: block;
        margin-bottom: 5px;
        margin-left: 20px;
        min-width: 80px;
    }
  language: css
libraries: |+
  https://appsforoffice.microsoft.com/lib/1/hosted/office.js
  @types/office-js

  https://cdn.jsdelivr.net/npm/@tensorflow/tfjs/dist/tf.min.js

  https://cdn.jsdelivr.net/npm/@tensorflow/tfjs-vis/dist/tfjs-vis.umd.min.js

  office-ui-fabric-js@1.4.0/dist/css/fabric.min.css
  office-ui-fabric-js@1.4.0/dist/css/fabric.components.min.css

  core-js@2.4.1/client/core.min.js
  @types/core-js

  jquery@3.1.1
  @types/jquery@3.3.1
	name: Basic API call (TypeScript)
	description: Performs a basic Excel API call using TypeScript.
	host: EXCEL
	api_set: {}
	script:
	content: \|
	$("#run").click(() => tryCatch(run));

	console.log("Ready");

	async function run() {
	const days = Number($("#days").val());
	const no_sims = Number($("#sims").val());
	const gencheck = $("#gencheck").is(":checked");
	const scale = Number($("#genscale").val()) / 100.0;
	const weekend_act = Number($("#weekend").val());
	const weekend = $("#wkdcheck").is(":checked");

	let schedule = [];

	if ($("#ccg").is(":checked")) {
	const ccgno = Number($("#ccgno").val());
	const ccgmin = Number($("#ccgmin").val());
	const ccgmax = Number($("#ccgmax").val());
	const ccglen = Number($("#ccglen").val());
	const ccgres = Number($("#ccgres").val());

	const e2 = Array.from(Array(ccgno).keys()).map((_) => ({
	func: spike,
	peak: () => sim_rand(ccgmin, ccgmax),
	length: () => sim_rand(1, ccglen),
	net: () => sim_rand(1, ccgres),
	when: () => sim_rand(1, days)
	}));
	schedule = schedule.concat(e2);
	}
	if ($("#ccg2").is(":checked")) {
	const ccg2no = Number($("#ccg2no").val());
	const ccg2txt = Number($("#ccg2txt").val());
	const ccg2conv = Number($("#ccg2conv").val());
	const ccg2len = Number($("#ccg2len").val());
	const ccg2res = Number($("#ccg2res").val());

	const e2 = Array.from(Array(ccg2no).keys()).map((_) => ({
	func: spike,
	peak: () => (ccg2conv / 100) * ccg2txt,
	length: () => sim_rand(2, ccg2len),
	net: () => ccg2res,
	when: () => sim_rand(1, days)
	}));
	schedule = schedule.concat(e2);
	}

	if ($("#covid").is(":checked")) {
	const cwhen = Number($("#cwhen").val());
	const cmin = Number($("#cmin").val());
	const cmax = Number($("#cmax").val());
	const cres = Number($("#cres").val());

	const e2 = [
	{
	func: spike,
	peak: () => sim_rand(cmin, cmax),
	length: () => 30,
	net: () => cres,
	when: () => cwhen
	}
	];
	schedule = schedule.concat(e2);
	}

	if ($("#covid2").is(":checked")) {
	const cwhen = Number($("#c2when").val());
	const cmin = Number($("#c2min").val());
	const cmax = Number($("#c2max").val());
	const cres = Number($("#c2res").val());

	const e2 = [
	{
	func: spike,
	peak: () => sim_rand(cmin, cmax),
	length: () => 30,
	net: () => cres,
	when: () => cwhen
	}
	];
	schedule = schedule.concat(e2);
	}

	if ($("#flu").is(":checked")) {
	const when = Number($("#wwhen").val());
	const len = Number($("#wlen").val());
	const extra = Number($("#wextra").val());
	const e2 = [
	{
	func: rise,
	days: () => len,
	net: () => extra,
	when: () => when
	}
	];
	schedule = schedule.concat(e2);
	}

	if ($("#bckgnd").is(":checked")) {
	const bfrom = Number($("#bfrm").val());
	const bperc = Number($("#bperc").val());
	const e2 = [
	{
	func: trend,
	c: () => bfrom,
	m: () => (bperc * (bfrom + 1)) / (100 * days),
	when: () => 0
	}
	];
	schedule = schedule.concat(e2);
	}

	if ($("#partners").is(":checked")) {
	const pnums = Number($("#pnums").val());
	const pscale = Number($("#pscale").val());
	const pperc = Number($("#pperc").val());

	const e2 = Array.from(Array(pnums).keys()).map(function(_) {
	let pfrom = pscale * 150;

	return {
	func: trend,
	c: () => pfrom,
	m: () => (pperc * (pfrom + 1)) / (100 * days),
	when: () => sim_rand(1, days)
	};
	});

	schedule = schedule.concat(e2);
	}

	console.log(schedule.length + " events scheduled");
	Excel.run(function(context) {
	console.log("Starting Sims");
	let sims = [];

	for (let i = 0; i < no_sims; i++) {
	const s = run_sim(schedule, weekend, weekend_act, scale, gencheck, days);
	const sync_s = s.arraySync();
	sims.push(sync_s);
	}
	console.log("Finished Sims");
	console.log("Calculate Statistics");

	const s2 = tf.tensor2d(sims);
	const s3 = tf.moments(s2, 0);
	const mean_s = s3["mean"].arraySync().map((x) => [x]);
	const cummean_s = s3["mean"]
	.cumsum()
	.arraySync()
	.map((x) => [x]);
	const stddev_s = s3["variance"]
	.sqrt()
	.arraySync()
	.map((x) => [x]);
	console.log("Updating working sheet");
	var sheet = context.workbook.worksheets.getItem("Simulations");

	var simrange = sheet.getRange("A1:A" + String(days + 1));
	simrange.values = [["Submissions: Sample Sim"]].concat(sims[0].map((x) => [x]));
	simrange.format.autofitColumns();

	var mrange = sheet.getRange("B1:B" + String(days + 1));
	mrange.values = [["Submissions: Mean of Sims"]].concat(mean_s);
	mrange.format.autofitColumns();

	var stdrange = sheet.getRange("C1:C" + String(days + 1));
	stdrange.values = [["Stddev"]].concat(stddev_s);
	stdrange.format.autofitColumns();

	var cumurange = sheet.getRange("D1:D" + String(days + 1));
	cumurange.values = [["Cumulative Submissions"]].concat(cummean_s);
	cumurange.format.autofitColumns();

	var chart1 = sheet.charts.add("Line", simrange, "auto");
	chart1.title.text = "Single Simulation: Submissions";

	var chart2 = sheet.charts.add("Line", mrange, "auto");
	chart2.title.text = "Mean of Simulations: Submissions";

	var chart3 = sheet.charts.add("Line", cumurange, "auto");
	chart3.title.text = "Mean of Simulations: Cumulative Submissions";

	chart1.setPosition("E1", "J16");
	chart2.setPosition("E18", "J32");
	chart3.setPosition("E34", "J48");
	console.log("finished");
	//chart.legend.position = "right"
	//chart.legend.format.fill.setSolidColor("white");
	//chart.dataLabels.format.font.size = 5;
	//chart.dataLabels.format.font.color = "black";

	return context.sync();
	});
	}
	function rise(ev, no_days = 365) {
	const days = ev["days"]();
	const net = ev["net"]();
	const when = ev["when"]();

	const y = tf.fill([days], net);
	return y.pad([[when, no_days - (when + days)]]);
	}

	function weekend_adjustment(no_days = 365, weekend) {
	//}, minScale = 0.6, maxScale = 0.7) {

	const minScale = weekend / 100 - 0.1;
	const maxScale = weekend / 100 + 0.1;
	const weeks = Math.floor(no_days / 7);
	const weekdays = tf.ones([weeks, 5]);
	const weekends = tf.randomUniform([weeks, 2], minScale, maxScale);
	const all_days = tf.concat([weekdays, weekends], (axis = 1)).flatten();
	const no_act_days = all_days.shape[0];
	return all_days.pad([[0, no_days - no_act_days]], 1);
	}

	function spike(ev, no_days = 365) {
	const peak = ev["peak"]();
	const len = ev["length"]();
	const net = ev["net"]();
	const when = ev["when"]();

	const l0 = (Math.log(peak) + 1) / len;
	const x = tf.range(0, len);
	const spike_out = tf.scalar(peak - net).mul(tf.exp(tf.scalar(-l0).mul(x)));
	const ele_spike_out = spike_out.add(tf.scalar(net));
	const start_padded_spike = ele_spike_out.pad([[when, 0]]);
	return start_padded_spike.pad([[0, no_days - (when + len)]], net);
	}

	function trend(ev, no_days = 365) {
	const m = ev["m"]();
	const c = ev["c"]();
	const when = ev["when"]();

	const x = tf.range(0, no_days);
	const trend_out = tf
	.scalar(m)
	.mul(x)
	.add(tf.scalar(c));
	return position(trend_out, when, no_days);
	}

	function position(data, when, no_days) {
	return data.pad([[when, no_days - (data.shape[0] + when)]]);
	}

	function run_sim(evts, wkdcheck, weekends, scale, gencheck, no_days = 365) {
	let yrs = tf.zeros([no_days]);
	for (var e in evts) {
	const event_days = evts[e]["func"](evts[e], no_days);
	yrs = yrs.add(event_days);
	}
	if (wkdcheck) {
	yrs = yrs.mul(weekend_adjustment(no_days, weekends));
	}
	if (gencheck) {
	yrs = tf.scalar(scale).mul(yrs);
	}
	return yrs;
	}

	function sim_rand(mn, mx) {
	const shell_out = (x) => x.arraySync()[0];
	return Math.floor(shell_out(tf.randomUniform([1], mn, mx)));
	}

	/** Default helper for invoking an action and handling errors. */
	async function tryCatch(callback) {
	try {
	await callback();
	} catch (error) {
	// Note: In a production add-in, you'd want to notify the user through your add-in's UI.
	console.error(error);
	}
	}
	language: typescript
	template:
	content: "<section class=\"samples ms-font-m\">\n\t<h1>NHS-Login Forecasting</h1>\n\t<p> Please fill in the details of the simulation below (it scrolls, and has 7 Scenario types) and press the\n\t\t<b>Generate Forecast</b> button.</p>\n\t<label> No. Simulations to aggregate: </label> <input id=\"sims\" type=\"text\" size=\"3\" value=\"10\"><br><br>\n\t<button id=\"run\" class=\"ms-Button\"> <span class=\"ms-Button-label\">Generate Forecast</span></button>\n\n\t<h2> Settings </h2>\n\n\t<h3> General </h3>\n\t<label> Length of forecast: </label>\n\t<input id=\"days\" type=\"text\" size=\"3\" value=\"365\"><label><em> (days)</em></label><br>\n\t<input type=\"checkbox\" id=\"wkdcheck\"> <label> Weekend as % of weekday: </label>\n\t<input id=\"weekend\" type=\"text\" size=\"3\" value=\"60\"><em> (1-100)+-10%</em><br>\n\t<input type=\"checkbox\" id=\"gencheck\"> <label> general scaling: </label>\n\t<input id=\"genscale\" type=\"text\" size=\"3\" value=\"100\"><em> (1-500)</em><br>\n\t<p> <em>General scaling </em> is a multiplying factor for the entire forecast. It's a way of upgrading/downgrading\n\t\tthe values to reflect things such as a more postive view of NHS-app numbers. </p>\n\n\t<h3> <label>1. Spike Events </label> <input type=\"checkbox\" id=\"ccg\"> </h3>\n\t\t<p><em> Spikes, randomly throughout period</em></p>\n\t\t<label> Number: </label> <input id=\"ccgno\" type=\"text\" size=\"3\" value=\"10\"><em> (events)</em><br>\n\t\t<label> Min Extra Submissions: </label>\n\t\t<input id=\"ccgmin\" type=\"text\" size=\"5\" value=\"4000\"><em> (persons)</em><br>\n\t\t<label> Max Extra Submissions: </label>\n\t\t<input id=\"ccgmax\" type=\"text\" size=\"5\" value=\"6000\"><em> (persons)</em><br>\n\t\t<label> Length: </label> <input id=\"ccglen\" type=\"text\" size=\"3\" value=\"2\"><em> (days)</em><br>\n\t\t<label> Residual after event up to: </label> <input id=\"ccgres\" type=\"text\" size=\"5\" value=\"500\">\n\t\t<em> (persons)</em>\n\n\t\t<h3> <label>2. CCG Events </label> <input type=\"checkbox\" id=\"ccg2\"> </h3>\n\t\t\t<p><em> Notts-style CCG event spikes, randomly throughout period</em></p>\n\t\t\t<label> Number: </label> <input id=\"ccg2no\" type=\"text\" size=\"3\" value=\"10\"><em> (events)</em><br>\n\t\t\t<label> Txts: </label>\n\t\t\t<input id=\"ccg2txt\" type=\"text\" size=\"6\" value=\"100000\"><em> (persons)</em><br>\n\t\t\t<label> % Conversion: </label>\n\t\t\t<input id=\"ccg2conv\" type=\"text\" size=\"3\" value=\"6\"><em> (1-100)</em><br>\n\t\t\t<label> Length: </label> <input id=\"ccg2len\" type=\"text\" size=\"3\" value=\"2\"><em> (days)</em><br>\n\t\t\t<label> Residual after event up to: </label> <input id=\"ccg2res\" type=\"text\" size=\"5\" value=\"500\">\n\t\t\t<em> (persons)</em>\n\n\t\t\t<h3> <label>3. Partners </label> <input type=\"checkbox\" id=\"partners\"> </h3>\n\t\t\t\t<p><em>How partners will grow over period. Assumption is linear growth. Co-op is \n\t\t\t\t\tone unit and equates to 150 submissions per day. We assume that these partners will have\n\t\t\t\t\ta significant impact as per co-op</em> </p>\n\t\t\t\t<label>Number of New significant partners in year: </label>\n\t\t\t\t<input id=\"pnums\" type=\"text\" size=\"3\" value=\"20\"><em> (companies)</em><br>\n\t\t\t\t<label>Max Company size when compared to the Co-op: </label>\n\t\t\t\t<input id=\"pscale\" type=\"text\" size=\"3\" value=\"1.0\"><em> </em><br>\n\t\t\t\t<label>% growth over year: </label> <input id=\"pperc\" type=\"text\" size=\"3\" value=\"10\"><em> (0-100)</em>\n\n\n\t\t\t\t<h3> <label>4. Background Growth </label> <input type=\"checkbox\" id=\"bckgnd\"> </h3>\n\t\t\t\t\t<p><em>General background growth over period. Assumption is linear growth. </em></p>\n\t\t\t\t\t<label>Starting value: </label>\n\t\t\t\t\t<input id=\"bfrm\" type=\"text\" size=\"5\" value=\"10000\"><em> (persons)</em><br>\n\t\t\t\t\t<label>% growth over year: </label>\n\t\t\t\t\t<input id=\"bperc\" type=\"text\" size=\"3\" value=\"10\"><em> (0-100)</em><br>\n\n\n\t\t\t\t\t<h3> <label>5. Covid Lockdown</label> <input type=\"checkbox\" id =\"covid\"> </h3>\n\t\t\t\t\t\t<p><em>A 1st major spikes at a period in the future</em></p>\n\t\t\t\t\t\t<label> When:</label> <input id=\"cwhen\" type=\"text\" size=\"3\" value=\"0\"><em> (day num.)</em><br>\n\t\t\t\t\t\t<label> Min Extra Submissions: </label>\n\t\t\t\t\t\t<input id=\"cmin\" type=\"text\" size=\"5\" value=\"18000\"><em> (persons)</em><br>\n\t\t\t\t\t\t<label> Max Extra Submissions:</label>\n\t\t\t\t\t\t<input id=\"cmax\" type=\"text\" size=\"5\" value=\"20000\"><em> (persons)</em><br>\n\t\t\t\t\t\t<label> Residual after event:</label>\n\t\t\t\t\t\t<input id=\"cres\" type=\"text\" size=\"5\" value=\"7500\"><em> (persons)</em><br>\n\n\t\t\t\t\t\t<h3> <label>6. Covid 2nd Lockdown</label> <input type=\"checkbox\" id =\"covid2\"> </h3>\n\t\t\t\t\t\t\t<p><em>A 2nd major spikes at a period in the future</em></p>\n\t\t\t\t\t\t\t<label> When:</label>\n\t\t\t\t\t\t\t<input id=\"c2when\" type=\"text\" size=\"3\" value=\"100\"><em> (day num.)</em><br>\n\t\t\t\t\t\t\t<label> Min Extra Submissions: </label>\n\t\t\t\t\t\t\t<input id=\"c2min\" type=\"text\" size=\"5\" value=\"18000\"><em> (persons)</em><br>\n\t\t\t\t\t\t\t<label> Max Extra Submissions:</label>\n\t\t\t\t\t\t\t<input id=\"c2max\" type=\"text\" size=\"5\" value=\"20000\"><em> (persons)</em><br>\n\t\t\t\t\t\t\t<label> Residual after event:</label>\n\t\t\t\t\t\t\t<input id=\"c2res\" type=\"text\" size=\"5\" value=\"7500\"><em> (persons)</em><br>\n\n\t\t\t\t\t\t\t<h3> <label>7. Winter Rise</label> <input type=\"checkbox\" id=\"flu\"> </h3>\n\t\t\t\t\t\t\t\t<p> <em>A general rise in usage typically associated with a november normal flu epidemic</em>\n\t\t\t\t\t\t\t\t</p>\n\t\t\t\t\t\t\t\t<label> When: </label>\n\t\t\t\t\t\t\t\t<input id=\"wwhen\" type=\"text\" size=\"3\" value=\"100\"><label><em> (day num.)</em></label><br>\n\t\t\t\t\t\t\t\t<label> For how long: </label>\n\t\t\t\t\t\t\t\t<input id=\"wlen\" type=\"text\" size=\"3\" value=\"50\"><em> (days)</em><br>\n\t\t\t\t\t\t\t\t<label> Extra Submissions: </label>\n\t\t\t\t\t\t\t\t<input id=\"wextra\" type=\"text\" size=\"5\" value=\"1000\"><em> (persons)</em><br><br>\n\n\n</section>"
	language: html
	style:
	content: \|-
	section.samples {
	margin-top: 20px;
	}

	section.samples .ms-Button, section.setup .ms-Button {
	display: block;
	margin-bottom: 5px;
	margin-left: 20px;
	min-width: 80px;
	}
	language: css
	libraries: \|+
	https://appsforoffice.microsoft.com/lib/1/hosted/office.js
	@types/office-js

	https://cdn.jsdelivr.net/npm/@tensorflow/tfjs/dist/tf.min.js

	https://cdn.jsdelivr.net/npm/@tensorflow/tfjs-vis/dist/tfjs-vis.umd.min.js

	office-ui-fabric-js@1.4.0/dist/css/fabric.min.css
	office-ui-fabric-js@1.4.0/dist/css/fabric.components.min.css

	core-js@2.4.1/client/core.min.js
	@types/core-js

	jquery@3.1.1
	@types/jquery@3.3.1