Compare commits
1
Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
|
b4a72a08a9 |
+11
-11
@@ -7,12 +7,12 @@ Written by `analysis/black/evidence.py` from `data/public/runs_clean.csv`. Count
|
||||
| Attempt | Form submissions | Bulk logs | Bot |
|
||||
|---|---|---|---|
|
||||
| 1 | 97 / 960 (10.1%) | 0 / 29 (0.0%) | 0 / 0 |
|
||||
| 2 | 0 / 299 (0.0%) | 0 / 21 (0.0%) | 0 / 4 (0.0%) |
|
||||
| 3 | 0 / 196 (0.0%) | 0 / 14 (0.0%) | 0 / 5 (0.0%) |
|
||||
| 4-10 | 0 / 555 (0.0%) | 0 / 102 (0.0%) | 0 / 36 (0.0%) |
|
||||
| 11-30 | 0 / 657 (0.0%) | 0 / 57 (0.0%) | 0 / 137 (0.0%) |
|
||||
| 31 or more | 0 / 715 (0.0%) | 0 / 0 | 0 / 333 (0.0%) |
|
||||
| not given | 0 / 1582 (0.0%) | 0 / 149 (0.0%) | 0 / 0 |
|
||||
| 2 | 0 / 296 (0.0%) | 0 / 21 (0.0%) | 0 / 4 (0.0%) |
|
||||
| 3 | 0 / 193 (0.0%) | 0 / 14 (0.0%) | 0 / 5 (0.0%) |
|
||||
| 4-10 | 0 / 538 (0.0%) | 0 / 102 (0.0%) | 0 / 36 (0.0%) |
|
||||
| 11-30 | 0 / 627 (0.0%) | 0 / 57 (0.0%) | 0 / 137 (0.0%) |
|
||||
| 31 or more | 0 / 694 (0.0%) | 0 / 0 | 0 / 333 (0.0%) |
|
||||
| not given | 0 / 1581 (0.0%) | 0 / 149 (0.0%) | 0 / 0 |
|
||||
|
||||
Complete logs (bot and bulk) have 29 first attempts at the tree and 709 later ones, with 0 shadows. Zero in 709 later attempts rules out a rate above 0.4% for them; zero in 29 first attempts only rules out more than 10%.
|
||||
|
||||
@@ -22,16 +22,16 @@ If the shadow depended only on the choices the form records (company, first dire
|
||||
|
||||
| Route | First attempt | Later attempts |
|
||||
|---|---|---|
|
||||
| Alone / West / Assess the situation / Engage the smaller cadet / Squadmate's revenge / West | 23 / 114 (20.2%) | 0 / 106 (0.0%) |
|
||||
| Alone / West / Assess the situation / Engage the smaller cadet / Squadmate's revenge / Forest | 5 / 29 (17.2%) | 0 / 96 (0.0%) |
|
||||
| Alone / West / Assess the situation / Engage the smaller cadet / Squadmate's revenge / West | 23 / 114 (20.2%) | 0 / 94 (0.0%) |
|
||||
| Alone / West / Assess the situation / Engage the smaller cadet / Squadmate's revenge / Forest | 5 / 29 (17.2%) | 0 / 95 (0.0%) |
|
||||
| Alone / Center / Assess the situation / Engage the smaller cadet / Squadmate's revenge / Forest | 4 / 29 (13.8%) | 0 / 78 (0.0%) |
|
||||
| 1 Squadmate / East / Assess the situation / Engage the smaller cadet / Squadmate's revenge / East | 2 / 32 (6.2%) | 0 / 73 (0.0%) |
|
||||
| 2 Squadmates / East / Assess the situation / Engage the smaller cadet / Squadmate's revenge / East | 2 / 45 (4.4%) | 0 / 50 (0.0%) |
|
||||
| 1 Squadmate / East / Assess the situation / Engage the smaller cadet / Squadmate's revenge / East | 2 / 32 (6.2%) | 0 / 67 (0.0%) |
|
||||
| Alone / East / Assess the situation / Engage the smaller cadet / Squadmate's revenge / East | 1 / 34 (2.9%) | 0 / 60 (0.0%) |
|
||||
| 2 Squadmates / East / Assess the situation / Engage the smaller cadet / Squadmate's revenge / East | 2 / 45 (4.4%) | 0 / 48 (0.0%) |
|
||||
| Alone / Center / Assess the situation / Daggers or not (protect squadmate) / Left bullies to the dragons / Forest | 1 / 28 (3.6%) | 0 / 64 (0.0%) |
|
||||
| Alone / Center / Assess the situation / Engage the smaller cadet / Squadmate's revenge / West | 2 / 9 (22.2%) | 0 / 81 (0.0%) |
|
||||
|
||||
Across all 326 recorded routes, comparing first and later attempts on the same route: 93 first-attempt shadows where 38.8 would be expected if the attempt did not matter (odds ratio inf, z = 12.5). First and later attempts choose almost the same routes, so the difference is not in the recorded choices: it is something about the account, or a choice the form does not record.
|
||||
Across all 325 recorded routes, comparing first and later attempts on the same route: 93 first-attempt shadows where 39.9 would be expected if the attempt did not matter (odds ratio inf, z = 12.3). First and later attempts choose almost the same routes, so the difference is not in the recorded choices: it is something about the account, or a choice the form does not record.
|
||||
|
||||
## 3. On one route, is the shadow decided by the cadet's wing, section and squad?
|
||||
|
||||
|
||||
+10
-10
@@ -93,8 +93,8 @@
|
||||
"description": "Every run from every source, one per row, coded into the same columns (collection/etl/runs.py; the form's rules are core/spec/form.yaml). The `source` column says where each run came from.",
|
||||
"contentUrl": "https://git.kenschiano.net/DragonKind/threshing-atlas/raw/branch/main/data/public/runs_clean.csv",
|
||||
"encodingFormat": "text/csv",
|
||||
"sha256": "e1cb9c0c0ab525c9fab393844ca00aaa6bbafa470c348f5035779c6b303b1137",
|
||||
"contentSize": "3720698 B"
|
||||
"sha256": "ce944619f3a8ca03e67e05438a630d632d6dbe10c13e557e9fcc2fa897cdc932",
|
||||
"contentSize": "3685124 B"
|
||||
},
|
||||
{
|
||||
"@type": "cr:FileObject",
|
||||
@@ -103,8 +103,8 @@
|
||||
"description": "The community \"Threshing Routes\" form export as published (headers and rows), with email addresses in free-text answers redacted. The source of runs_clean.",
|
||||
"contentUrl": "https://git.kenschiano.net/DragonKind/threshing-atlas/raw/branch/main/data/public/form_responses.json",
|
||||
"encodingFormat": "application/json",
|
||||
"sha256": "8067ac944db2833e6a1203dc6464ed208d60dea4e617ee9b28fc56f285445d2e",
|
||||
"contentSize": "12786245 B"
|
||||
"sha256": "3a3322da84274416880d87a400b06e44e0ed97feaf93df839b5269bdaa1badad",
|
||||
"contentSize": "12640700 B"
|
||||
},
|
||||
{
|
||||
"@type": "cr:FileObject",
|
||||
@@ -143,8 +143,8 @@
|
||||
"description": "Every number the Atlas page shows, for six filter presets (analysis/stats/site_stats.py; what is measured is analysis/stats/measures.yaml).",
|
||||
"contentUrl": "https://git.kenschiano.net/DragonKind/threshing-atlas/raw/branch/main/data/public/stats.json",
|
||||
"encodingFormat": "application/json",
|
||||
"sha256": "01047d7f9404d689eb688d04d6473d4705de84b112f0671464d4bcef46a49596",
|
||||
"contentSize": "5496605 B"
|
||||
"sha256": "d2a551e785c8593b442d95338a99ea0710ed7df2cd6c6f3ef341bddff1f3689e",
|
||||
"contentSize": "5475502 B"
|
||||
},
|
||||
{
|
||||
"@type": "cr:FileObject",
|
||||
@@ -153,8 +153,8 @@
|
||||
"description": "The route model: the game (core/spec/game.yaml), the count tables behind every roll, and the best decision tree for every dragon colour under two goals, with posterior intervals (planner/build_policy.py).",
|
||||
"contentUrl": "https://git.kenschiano.net/DragonKind/threshing-atlas/raw/branch/main/data/public/policy.json",
|
||||
"encodingFormat": "application/json",
|
||||
"sha256": "f65aaebb23cb3f56dad0387a94113857eb44c4ba077cfe5fc6215c202546588c",
|
||||
"contentSize": "987411 B"
|
||||
"sha256": "ef607b9a7fbc7f7d0e647f988fbac94a1685109a78218ce8c6f0ada3e62e2ee4",
|
||||
"contentSize": "986552 B"
|
||||
},
|
||||
{
|
||||
"@type": "cr:FileObject",
|
||||
@@ -163,8 +163,8 @@
|
||||
"description": "The route graph over every run in runs_clean.csv, counted per source (analysis/network/build-graph.js).",
|
||||
"contentUrl": "https://git.kenschiano.net/DragonKind/threshing-atlas/raw/branch/main/data/public/graph.json",
|
||||
"encodingFormat": "application/json",
|
||||
"sha256": "3b11d237468198458453a6066d69d85f5b3bef99e58241bb83a4ab313e3e2969",
|
||||
"contentSize": "75548 B"
|
||||
"sha256": "112895b711ade4e0aad8e4b7a61dea5db00f4734e484bc20198740e924c14974",
|
||||
"contentSize": "75580 B"
|
||||
}
|
||||
],
|
||||
"recordSet": [
|
||||
|
||||
+5
-5
@@ -10,18 +10,18 @@
|
||||
"changed": "2026-11-16"
|
||||
},
|
||||
"reddit_spreadsheet": {
|
||||
"sha256": "44eeece5fd75fb27f0c7bdf2729d812ba710f4bfba30dfd5dbee51a7888b8270",
|
||||
"bytes": 5994979,
|
||||
"sha256": "ca06bbb6914483214801b54b58ed08039c21dc858c59347456ea473b1417fbb4",
|
||||
"bytes": 5941127,
|
||||
"changed": "2026-11-19"
|
||||
},
|
||||
"register": {
|
||||
"sha256": "6efdf107fce4d7b5a48de3d683762b7236a6049c875145bf4aa1608c90f9ae8f",
|
||||
"sha256": "d6e5085f73adbf047163578c150dd034e2387fe37bd91030b36592df46147261",
|
||||
"bytes": 51331,
|
||||
"changed": "2026-11-19"
|
||||
},
|
||||
"reported": {
|
||||
"sha256": "4c62d2e726e7c967d03a6c85d3bb3e09384656b7ffd9ea629d13400fdc37b211",
|
||||
"bytes": 28296,
|
||||
"sha256": "8c8721b9c698e155c406da97fb69dda1729a9546290773dc0a044b1cd7eae628",
|
||||
"bytes": 27492,
|
||||
"changed": "2026-11-19"
|
||||
}
|
||||
}
|
||||
|
||||
+10
-10
@@ -8,23 +8,23 @@
|
||||
],
|
||||
"2": [
|
||||
0,
|
||||
299
|
||||
296
|
||||
],
|
||||
"3": [
|
||||
0,
|
||||
196
|
||||
193
|
||||
],
|
||||
"4-10": [
|
||||
0,
|
||||
555
|
||||
538
|
||||
],
|
||||
"11-30": [
|
||||
0,
|
||||
657
|
||||
627
|
||||
],
|
||||
"31 or more": [
|
||||
0,
|
||||
715
|
||||
694
|
||||
]
|
||||
},
|
||||
"bulk logs": {
|
||||
@@ -81,11 +81,11 @@
|
||||
}
|
||||
},
|
||||
"same_route": {
|
||||
"routes": 326,
|
||||
"routes": 325,
|
||||
"first_shadows": 93,
|
||||
"expected": 38.8,
|
||||
"expected": 39.9,
|
||||
"odds_ratio": Infinity,
|
||||
"z": 12.5
|
||||
"z": 12.3
|
||||
},
|
||||
"form_first": [
|
||||
97,
|
||||
@@ -93,7 +93,7 @@
|
||||
],
|
||||
"form_later": [
|
||||
0,
|
||||
2422
|
||||
2348
|
||||
],
|
||||
"best_route": {
|
||||
"route": [
|
||||
@@ -110,7 +110,7 @@
|
||||
],
|
||||
"later": [
|
||||
0,
|
||||
106
|
||||
94
|
||||
]
|
||||
},
|
||||
"shadow_routes": {
|
||||
|
||||
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
File diff suppressed because one or more lines are too long
+2361
-2471
File diff suppressed because it is too large
Load Diff
File diff suppressed because one or more lines are too long
+35
-1
@@ -753,6 +753,37 @@
|
||||
${trafficToggle()}`;
|
||||
}
|
||||
|
||||
/** Inside the bonding encounter: the answers the bot gave, screen by screen, on its best logged way to bond this
|
||||
* colour (data/tree.json, analysis/tree/build_tree.py), with what every other answer did. Only shown when the bot's
|
||||
* way goes through the same encounter as the route; the community form does not record these answers. */
|
||||
const TREE_SIDE = (enc) => (/^(valley|meadow)-/.test(enc) ? "valley" : "tree");
|
||||
function encounterAnswers(enc) {
|
||||
const t = st.tree;
|
||||
if (!t) return '<p class="small muted ex-inside">Loading the answers inside the encounter…</p>';
|
||||
const best = (t.best || {})[st.colour];
|
||||
const picks = best && best.picks && best.picks.length ? best.picks : null;
|
||||
const hourAt = picks ? picks.findIndex((p) => (t.screens[p.screen] || {}).stage === "THORN_WALK") : -1;
|
||||
const side = hourAt >= 0 && /valley/i.test(picks[hourAt].text) ? "valley" : "tree";
|
||||
const viaShadow = picks && picks.some((p) => /shadow/i.test(p.text) && !/descend/i.test(p.text));
|
||||
const fits = picks && hourAt >= 0 && side === TREE_SIDE(enc) && (enc.startsWith("shadow-") ? viaShadow : !viaShadow || side === "valley");
|
||||
if (!fits) return `<p class="small muted ex-inside">The bot hasn't bonded a ${st.colour} through ${esc(ENC_NAME[enc] || enc)} yet, so the answers inside this encounter aren't mapped. The community form records only how the encounter ended, not the answers given.</p>`;
|
||||
const died = (o) => (o.to && o.to.DIED ? o.to.DIED.n : 0);
|
||||
const rows = picks.slice(hourAt + 1).map((p, i) => {
|
||||
const sc = t.screens[p.screen] || { options: {} };
|
||||
const me = sc.options[p.option] || { n: p.n, to: {} };
|
||||
const lived = me.n - died(me);
|
||||
const others = Object.entries(sc.options).filter(([k]) => k !== p.option).map(([, o]) =>
|
||||
`<li>${esc(o.text)} <small>${o.n ? `died ${died(o)} of ${o.n}` : "untested"}</small></li>`).join("");
|
||||
return `<li><span class="n">${i + 1}</span><div><b>${esc(p.text)}</b>
|
||||
<small>${me.n ? `got through ${lived} of ${me.n} times` : ""}${me.kind === "fixed" && me.n ? " · always the same so far" : me.kind === "dice" ? " · a roll" : ""}</small>
|
||||
${others ? `<details><summary>Other answers here</summary><ul>${others}</ul></details>` : ""}</div></li>`;
|
||||
}).join("");
|
||||
const end = Object.keys((t.screens[picks[picks.length - 1].screen] || { options: {} }).options[picks[picks.length - 1].option]?.to || {}).find((k) => k.startsWith("BONDED"));
|
||||
return `<div class="ex-inside"><span class="eyebrow">Inside the encounter: the answers that bonded</span>
|
||||
<ol>${rows}</ol>
|
||||
<p class="small muted">${end ? `Bonded a ${end.split(":")[1]}. ` : ""}From the bot's own runs (it logs every screen it sees), so the counts are small; untested answers may also work.</p></div>`;
|
||||
}
|
||||
|
||||
function panelRoutes(routes) {
|
||||
const best = Math.max(1e-12, ...routes.map((r) => r.value.x));
|
||||
const T = model().targets[st.colour], b = T[st.objective];
|
||||
@@ -772,7 +803,8 @@
|
||||
const open = st.expanded === s.var;
|
||||
const opts = open ? `<div class="ex-opts">${Object.entries(s.options).sort((a, c) => c[1].pX - a[1].pX).map(([v, val]) =>
|
||||
`<button type="button" class="ex-opt${v === s.taken ? " on" : ""}" data-var="${s.var}" data-v="${esc(v)}"><span>${esc(v)}${v === s.choice ? ' <em class="rec">best</em>' : ""}</span><b>${pct(val.pX)}</b></button>`).join("")}</div>` : "";
|
||||
return `<li class="dec${open ? " open" : ""}"><button type="button" class="ex-step" data-var="${s.var}" aria-expanded="${open}"><span>${esc(s.label)}</span><b>${esc(String(s.taken))}</b><small>${s.userSet ? "your choice" : "best"} · ${s.var === "encounter" ? pct(s.options[s.taken].pX) : pct(s.options[s.taken].pX)}</small></button>${opts}</li>`;
|
||||
const inside = open && s.var === "encounter" ? encounterAnswers(focus.enc) : "";
|
||||
return `<li class="dec${open ? " open" : ""}"><button type="button" class="ex-step" data-var="${s.var}" aria-expanded="${open}"><span>${esc(s.label)}</span><b>${esc(String(s.taken))}</b><small>${s.userSet ? "your choice" : "best"} · ${s.var === "encounter" ? pct(s.options[s.taken].pX) : pct(s.options[s.taken].pX)}</small></button>${opts}${inside}</li>`;
|
||||
}).join("");
|
||||
const risks = focus.steps.filter((s) => s.risk && s.risk.p > 0.0005).map((s) => `${s.risk.label.toLowerCase()} ${pct(s.risk.p)}`);
|
||||
const lb = st.policy.samples.lowBias.targets[st.colour].thisAttempt;
|
||||
@@ -926,6 +958,8 @@
|
||||
|
||||
function init() {
|
||||
render();
|
||||
// the screen tree (the answers inside each encounter), for the "Bonding encounter" step
|
||||
fetch("data/tree.json").then((r) => r.json()).then((t) => { st.tree = t; if (st.expanded === "encounter") render(); }).catch(() => { st.tree = { best: {}, screens: {} }; });
|
||||
fetch("data/policy.json").then((r) => r.json()).then((p) => {
|
||||
st.policy = p;
|
||||
S.useGame(p.samples.all); // the game (and where each colour can be bonded) comes with the route model
|
||||
|
||||
@@ -346,6 +346,14 @@ footer { border-top-color: var(--line); }
|
||||
.ex-kinds { display: flex; flex-wrap: wrap; gap: 4px 14px; margin-top: 4px; font-size: var(--step--1); color: var(--muted); }
|
||||
.ex-kind { display: inline-flex; align-items: center; gap: 6px; }
|
||||
.ex-kind svg { width: 34px; height: 8px; }
|
||||
.ex-inside { margin: 10px 0 4px; padding: 10px 12px; border-left: 2px solid var(--c, var(--accent)); background: color-mix(in srgb, var(--bg) 70%, transparent); border-radius: 0 8px 8px 0; }
|
||||
.ex-inside ol { list-style: none; margin: 6px 0; padding: 0; display: grid; gap: 8px; }
|
||||
.ex-inside ol > li { display: grid; grid-template-columns: 22px minmax(0, 1fr); gap: 8px; align-items: start; }
|
||||
.ex-inside .n { font-family: var(--font-display); color: var(--muted); text-align: right; }
|
||||
.ex-inside b { display: block; font-weight: 600; }
|
||||
.ex-inside small { color: var(--muted); }
|
||||
.ex-inside details summary { cursor: pointer; font-size: var(--step--1); color: var(--muted); }
|
||||
.ex-inside ul { margin: 4px 0 0 16px; padding: 0; font-size: var(--step--1); }
|
||||
.ex-head { display: grid; grid-template-columns: 64px minmax(0, 1fr) auto; gap: 10px; align-items: center; }
|
||||
.ex-head svg { width: 64px; height: 52px; }
|
||||
.ex-head h2 { color: var(--c); }
|
||||
|
||||
Reference in New Issue
Block a user