Readiness counts deepening: the mastered share sets where it starts (#15)
Browse filesThe share of the test's skills mastered closes 20% of the gap from 0; tests then pull the number up from there.
- docs/tasks/prd-mathly.md +2 -2
- mathly2/readiness.py +21 -10
- mathly2/realtests.py +8 -3
- mathly2/static/testcard.js +6 -5
- mathly2/tests/test_home_browser.py +1 -1
- mathly2/tests/test_readiness.py +11 -1
- mathly2/tests/test_realtests_api.py +19 -5
- mathly2/tests/test_writtentest_api.py +1 -1
- openspec/changes/readiness-counts-deepening/proposal.md +24 -0
- openspec/changes/readiness-counts-deepening/specs/readiness-meter/spec.md +12 -0
- openspec/changes/readiness-counts-deepening/tasks.md +5 -0
docs/tasks/prd-mathly.md
CHANGED
|
@@ -335,13 +335,13 @@ The exercise screen is where a student practices one skill.
|
|
| 335 |
- **GRD-2 When time ends.** Only the simulation is timed. When its timer ends it is submitted as it stands, and unanswered questions count as wrong.
|
| 336 |
Leaving the screen does not stop the timer.
|
| 337 |
- **RDY-1 Readiness meter.** For each upcoming test, the system shows the student (and the teacher) one **readiness number from 0 to 100**, with its level and a bar. Levels: 0–39 "מתחילים", 40–69 "בדרך", 70–84 "כמעט שם", 85–100 "מוכנים". Beside it, the app shows the points to the next level.
|
| 338 |
-
- **RDY-2
|
| 339 |
|
| 340 |
`number = number + GAIN × (grade − number)`
|
| 341 |
|
| 342 |
GAIN is 0.6 for a simulation, 0.3 for a practice test and 0.15 for a manual practice test.
|
| 343 |
- **RDY-3 Never down, never above the grades.** A grade at or below the number changes nothing, so the number never drops, and it can only approach the grades actually earned.
|
| 344 |
-
- **RDY-4 The number explains itself.** The test's page says how to raise it: a practice test closes 30% of the gap, a simulation 60%, a manual one 15%; a lower grade does not lower it
|
| 345 |
- **RDY-5 Make raising it rewarding.**
|
| 346 |
- The number counts up from the last visit with a "+N" badge.
|
| 347 |
- Every button to a test shows the most it can add now.
|
|
|
|
| 335 |
- **GRD-2 When time ends.** Only the simulation is timed. When its timer ends it is submitted as it stands, and unanswered questions count as wrong.
|
| 336 |
Leaving the screen does not stop the timer.
|
| 337 |
- **RDY-1 Readiness meter.** For each upcoming test, the system shows the student (and the teacher) one **readiness number from 0 to 100**, with its level and a bar. Levels: 0–39 "מתחילים", 40–69 "בדרך", 70–84 "כמעט שם", 85–100 "מוכנים". Beside it, the app shows the points to the next level.
|
| 338 |
+
- **RDY-2 Deepening, then tests.** Deepening sets where the number starts: the share of the test's skills mastered (0–100) closes 20% of the gap from 0, so all of the material mastered starts it at 20. Then each handed-in practice test or simulation, oldest first, when its grade is above the number:
|
| 339 |
|
| 340 |
`number = number + GAIN × (grade − number)`
|
| 341 |
|
| 342 |
GAIN is 0.6 for a simulation, 0.3 for a practice test and 0.15 for a manual practice test.
|
| 343 |
- **RDY-3 Never down, never above the grades.** A grade at or below the number changes nothing, so the number never drops, and it can only approach the grades actually earned.
|
| 344 |
+
- **RDY-4 The number explains itself.** The test's page says how to raise it: mastering the test's skills in deepening raises the start (up to 20, with how much it gives now); a practice test closes 30% of the gap, a simulation 60%, a manual one 15%; a lower grade does not lower it.
|
| 345 |
- **RDY-5 Make raising it rewarding.**
|
| 346 |
- The number counts up from the last visit with a "+N" badge.
|
| 347 |
- Every button to a test shows the most it can add now.
|
mathly2/readiness.py
CHANGED
|
@@ -1,14 +1,17 @@
|
|
| 1 |
-
"""The readiness number for a real test: one 0-100 number
|
|
|
|
| 2 |
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
|
|
|
| 6 |
|
| 7 |
number' = number + GAIN[kind] * (grade - number) when grade > number
|
| 8 |
"""
|
| 9 |
from __future__ import annotations
|
| 10 |
|
| 11 |
GAIN = {"simulation": 0.6, "practice": 0.3, "manual": 0.15}
|
|
|
|
| 12 |
LEVELS = [(0, "מתחילים"), (40, "בדרך"), (70, "כמעט שם"), (85, "מוכנים")]
|
| 13 |
|
| 14 |
|
|
@@ -26,21 +29,29 @@ def step(value: float, kind: str, grade: float) -> float:
|
|
| 26 |
return value + GAIN[kind] * (grade - value) if grade > value else value
|
| 27 |
|
| 28 |
|
| 29 |
-
def
|
| 30 |
-
"""
|
| 31 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 32 |
for kind, grade in results:
|
| 33 |
value = step(value, kind, grade)
|
| 34 |
number = max(0, min(100, int(value + 0.5)))
|
| 35 |
best = lambda kinds: max((g for k, g in results if k in kinds), default=None)
|
| 36 |
-
return {"number": number, "label": label(number), "next": next_level(number),
|
|
|
|
| 37 |
"practices": sum(1 for k, _ in results if k != "simulation"), "simulations": sum(1 for k, _ in results if k == "simulation"),
|
| 38 |
"practice_best": best(("practice", "manual")), "simulation_best": best(("simulation",)),
|
| 39 |
"can_add": {k: max(0, int(GAIN[k] * (100 - value) + 0.5)) for k in GAIN}} # what a perfect test of each kind would add now
|
| 40 |
|
| 41 |
|
| 42 |
-
def change(before: list[tuple[str, int]], result: tuple[str, int]) -> dict:
|
| 43 |
"""What one handed-in test did to the number: before, after, the gain, and whether it reached a new level."""
|
| 44 |
-
old, new = compute(before), compute([*before, result])
|
| 45 |
return {"before": old["number"], "after": new["number"], "gain": new["number"] - old["number"], "label": new["label"],
|
| 46 |
"level_up": new["label"] != old["label"], "next": new["next"]}
|
|
|
|
| 1 |
+
"""The readiness number for a real test: one 0-100 number moved by deepening (the test's skills mastered in the path) and by tests
|
| 2 |
+
(practice tests and simulations).
|
| 3 |
|
| 4 |
+
Deepening comes first, as the starting point: the share of the test's skills mastered (0-100) closes 20% of the gap from 0. Then every
|
| 5 |
+
handed-in test, oldest first, pulls the number toward its grade, and only up: a grade above the number closes part of the gap, a grade at
|
| 6 |
+
or below it changes nothing. A simulation closes more of the gap than a practice test; a practice test on skills the student picked closes
|
| 7 |
+
less (it can skip the hard ones). Mastering more skills raises the start, and so the number, whatever tests came after.
|
| 8 |
|
| 9 |
number' = number + GAIN[kind] * (grade - number) when grade > number
|
| 10 |
"""
|
| 11 |
from __future__ import annotations
|
| 12 |
|
| 13 |
GAIN = {"simulation": 0.6, "practice": 0.3, "manual": 0.15}
|
| 14 |
+
DEEPENING_GAIN = 0.2 # the share of the test's skills mastered pulls the start this much
|
| 15 |
LEVELS = [(0, "מתחילים"), (40, "בדרך"), (70, "כמעט שם"), (85, "מוכנים")]
|
| 16 |
|
| 17 |
|
|
|
|
| 29 |
return value + GAIN[kind] * (grade - value) if grade > value else value
|
| 30 |
|
| 31 |
|
| 32 |
+
def start(mastered_share: float | None) -> float:
|
| 33 |
+
"""The number before any test: 20% of the share of the test's skills mastered (0-1)."""
|
| 34 |
+
return DEEPENING_GAIN * 100 * max(0.0, min(1.0, mastered_share or 0.0))
|
| 35 |
+
|
| 36 |
+
|
| 37 |
+
def compute(results: list[tuple[str, int]], mastered_share: float | None = None) -> dict:
|
| 38 |
+
"""`results`: the handed-in tests of one real test, oldest first, as (kind, grade) with kind simulation, practice or manual.
|
| 39 |
+
`mastered_share`: the share (0-1) of the test's skills mastered in deepening."""
|
| 40 |
+
value = start(mastered_share)
|
| 41 |
+
deepening = int(value + 0.5)
|
| 42 |
for kind, grade in results:
|
| 43 |
value = step(value, kind, grade)
|
| 44 |
number = max(0, min(100, int(value + 0.5)))
|
| 45 |
best = lambda kinds: max((g for k, g in results if k in kinds), default=None)
|
| 46 |
+
return {"number": number, "label": label(number), "next": next_level(number), "deepening": deepening,
|
| 47 |
+
"mastered_share": round(max(0.0, min(1.0, mastered_share or 0.0)), 3),
|
| 48 |
"practices": sum(1 for k, _ in results if k != "simulation"), "simulations": sum(1 for k, _ in results if k == "simulation"),
|
| 49 |
"practice_best": best(("practice", "manual")), "simulation_best": best(("simulation",)),
|
| 50 |
"can_add": {k: max(0, int(GAIN[k] * (100 - value) + 0.5)) for k in GAIN}} # what a perfect test of each kind would add now
|
| 51 |
|
| 52 |
|
| 53 |
+
def change(before: list[tuple[str, int]], result: tuple[str, int], mastered_share: float | None = None) -> dict:
|
| 54 |
"""What one handed-in test did to the number: before, after, the gain, and whether it reached a new level."""
|
| 55 |
+
old, new = compute(before, mastered_share), compute([*before, result], mastered_share)
|
| 56 |
return {"before": old["number"], "after": new["number"], "gain": new["number"] - old["number"], "label": new["label"],
|
| 57 |
"level_up": new["label"] != old["label"], "next": new["next"]}
|
mathly2/realtests.py
CHANGED
|
@@ -109,8 +109,13 @@ def _result(request: Request, row: dict) -> dict:
|
|
| 109 |
return attempts.result(request.app.state.bank, row)
|
| 110 |
|
| 111 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 112 |
def readiness_for(facts: dict) -> dict:
|
| 113 |
-
return readiness.compute(facts["results"])
|
| 114 |
|
| 115 |
|
| 116 |
def readiness_change(request: Request, row: dict) -> dict | None:
|
|
@@ -123,7 +128,7 @@ def readiness_change(request: Request, row: dict) -> dict | None:
|
|
| 123 |
return None
|
| 124 |
facts = facts_for(request, row["user_id"], test, clock.school_day())
|
| 125 |
before = [(_kind_of(r), r["grade"]) for r in facts["graded"] if r["submitted_at"] < row["submitted_at"]]
|
| 126 |
-
return readiness.change(before, (_kind_of(row), row["grade"]))
|
| 127 |
|
| 128 |
|
| 129 |
def _title(row: dict) -> str:
|
|
@@ -145,7 +150,7 @@ def card(request: Request, user_id: str | None, test: dict, today: date) -> dict
|
|
| 145 |
trail.append({"attempt": row["id"], "kind": row["kind"], "title": _title(row), "grade": None, "checking": True, "gain": 0,
|
| 146 |
"number": None, "after": None, "at": row["submitted_at"].isoformat()})
|
| 147 |
continue
|
| 148 |
-
moved = readiness.change(before, (_kind_of(row), row["grade"]))
|
| 149 |
before.append((_kind_of(row), row["grade"]))
|
| 150 |
trail.append({"attempt": row["id"], "kind": row["kind"], "title": _title(row), "grade": row["grade"], "gain": moved["gain"],
|
| 151 |
"number": int(row["scope"].split(":")[1]) if row["kind"] == "exercise_test" else None,
|
|
|
|
| 109 |
return attempts.result(request.app.state.bank, row)
|
| 110 |
|
| 111 |
|
| 112 |
+
def mastered_share(facts: dict) -> float:
|
| 113 |
+
"""The share of the test's skills mastered (deepening counts toward readiness)."""
|
| 114 |
+
return facts["mastered"] / len(facts["covered"]) if facts["covered"] else 0.0
|
| 115 |
+
|
| 116 |
+
|
| 117 |
def readiness_for(facts: dict) -> dict:
|
| 118 |
+
return readiness.compute(facts["results"], mastered_share(facts))
|
| 119 |
|
| 120 |
|
| 121 |
def readiness_change(request: Request, row: dict) -> dict | None:
|
|
|
|
| 128 |
return None
|
| 129 |
facts = facts_for(request, row["user_id"], test, clock.school_day())
|
| 130 |
before = [(_kind_of(r), r["grade"]) for r in facts["graded"] if r["submitted_at"] < row["submitted_at"]]
|
| 131 |
+
return readiness.change(before, (_kind_of(row), row["grade"]), mastered_share(facts))
|
| 132 |
|
| 133 |
|
| 134 |
def _title(row: dict) -> str:
|
|
|
|
| 150 |
trail.append({"attempt": row["id"], "kind": row["kind"], "title": _title(row), "grade": None, "checking": True, "gain": 0,
|
| 151 |
"number": None, "after": None, "at": row["submitted_at"].isoformat()})
|
| 152 |
continue
|
| 153 |
+
moved = readiness.change(before, (_kind_of(row), row["grade"]), mastered_share(facts))
|
| 154 |
before.append((_kind_of(row), row["grade"]))
|
| 155 |
trail.append({"attempt": row["id"], "kind": row["kind"], "title": _title(row), "grade": row["grade"], "gain": moved["gain"],
|
| 156 |
"number": int(row["scope"].split(":")[1]) if row["kind"] == "exercise_test" else None,
|
mathly2/static/testcard.js
CHANGED
|
@@ -13,11 +13,12 @@ const MODES = [
|
|
| 13 |
];
|
| 14 |
const STATE = { mastered: "👑", learning: "◐", not_started: "○" };
|
| 15 |
|
| 16 |
-
/** How the number moves, in plain words. */
|
| 17 |
-
const
|
|
|
|
|
|
|
| 18 |
<li>כל מבחן שהציון בו גבוה מהמוכנות מעלה אותה: מבחן תרגול סוגר 30% מהפער, <b>מבחן סימולציה סוגר 60%</b>, ומבחן בבחירה ידנית 15%.</li>
|
| 19 |
-
<li>ציון נמוך מהמוכנות לא מוריד אותה.</li>
|
| 20 |
-
<li>רק מבחנים נספרים. תרגול במסלול ובהעמקה עוזר להצליח בהם, אבל לא משנה את המספר.</li></ul></details>`;
|
| 21 |
|
| 22 |
function simulationCard(test) {
|
| 23 |
const s = test.simulation, gain = test.readiness.can_add.simulation;
|
|
@@ -95,7 +96,7 @@ export async function show(view, id, focus = "") {
|
|
| 95 |
view(`${tabs("home")}<p><a class="back" href="${homeHref()}">‹ הבית</a></p>
|
| 96 |
<section class="card" id="test-card" data-test="${test.id}"><h2>מבחן ב־${shortDate(test.date)} <span class="muted">(${daysText(test.days_left)})</span></h2>
|
| 97 |
<p class="muted">${test.lessons.map(l => fmt(l.name)).join(", ")}${test.note ? ` · ${esc(test.note)}` : ""}</p>
|
| 98 |
-
${readinessMeter(test)}${
|
| 99 |
${simulationCard(test)}
|
| 100 |
${practiceCard(test)}
|
| 101 |
${trailCard(test)}
|
|
|
|
| 13 |
];
|
| 14 |
const STATE = { mastered: "👑", learning: "◐", not_started: "○" };
|
| 15 |
|
| 16 |
+
/** How the number moves, in plain words, with what deepening gives now. */
|
| 17 |
+
const how = r => `<details class="rd-how" id="rd-how"><summary>איך מעלים את המוכנות?</summary><ul>
|
| 18 |
+
<li id="rd-deepening">העמקה נספרת: כל מיומנות מהחומר של המבחן שמגיעים בה לשליטה (👑) מעלה את נקודת ההתחלה, עד 20 נקודות כשכל החומר בשליטה.
|
| 19 |
+
${r.mastered_share ? `<b>עכשיו: ${Math.round(r.mastered_share * 100)}% מהמיומנויות בשליטה, ${r.deepening} נקודות.</b>` : ""}</li>
|
| 20 |
<li>כל מבחן שהציון בו גבוה מהמוכנות מעלה אותה: מבחן תרגול סוגר 30% מהפער, <b>מבחן סימולציה סוגר 60%</b>, ומבחן בבחירה ידנית 15%.</li>
|
| 21 |
+
<li>ציון נמוך מהמוכנות לא מוריד אותה.</li></ul></details>`;
|
|
|
|
| 22 |
|
| 23 |
function simulationCard(test) {
|
| 24 |
const s = test.simulation, gain = test.readiness.can_add.simulation;
|
|
|
|
| 96 |
view(`${tabs("home")}<p><a class="back" href="${homeHref()}">‹ הבית</a></p>
|
| 97 |
<section class="card" id="test-card" data-test="${test.id}"><h2>מבחן ב־${shortDate(test.date)} <span class="muted">(${daysText(test.days_left)})</span></h2>
|
| 98 |
<p class="muted">${test.lessons.map(l => fmt(l.name)).join(", ")}${test.note ? ` · ${esc(test.note)}` : ""}</p>
|
| 99 |
+
${readinessMeter(test)}${how(test.readiness)}</section>
|
| 100 |
${simulationCard(test)}
|
| 101 |
${practiceCard(test)}
|
| 102 |
${trailCard(test)}
|
mathly2/tests/test_home_browser.py
CHANGED
|
@@ -489,7 +489,7 @@ def test_the_tests_page_has_three_practice_modes_and_the_trail_of_gains(browser,
|
|
| 489 |
page.click(".mx-x")
|
| 490 |
page.wait_for_selector("#practice")
|
| 491 |
page.click("#rd-how summary")
|
| 492 |
-
assert "סימולציה סוגר 60%" in page.inner_text("#rd-how") and page.errors == []
|
| 493 |
no_sideways_scroll(page)
|
| 494 |
|
| 495 |
|
|
|
|
| 489 |
page.click(".mx-x")
|
| 490 |
page.wait_for_selector("#practice")
|
| 491 |
page.click("#rd-how summary")
|
| 492 |
+
assert "סימולציה סוגר 60%" in page.inner_text("#rd-how") and "העמקה נספרת" in page.inner_text("#rd-deepening") and page.errors == []
|
| 493 |
no_sideways_scroll(page)
|
| 494 |
|
| 495 |
|
mathly2/tests/test_readiness.py
CHANGED
|
@@ -1,4 +1,4 @@
|
|
| 1 |
-
"""The readiness number:
|
| 2 |
from mathly2 import readiness
|
| 3 |
|
| 4 |
|
|
@@ -28,3 +28,13 @@ def test_change_says_the_gain_and_a_new_level():
|
|
| 28 |
moved = readiness.change([("practice", 100)], ("simulation", 90))
|
| 29 |
assert moved == {**moved, "before": 30, "after": 66, "gain": 36, "label": "בדרך", "level_up": True}
|
| 30 |
assert readiness.change([("simulation", 90)], ("practice", 20))["gain"] == 0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
"""The readiness number: deepening sets where it starts; tests move it only up, a simulation more than a practice test, a picked one less."""
|
| 2 |
from mathly2 import readiness
|
| 3 |
|
| 4 |
|
|
|
|
| 28 |
moved = readiness.change([("practice", 100)], ("simulation", 90))
|
| 29 |
assert moved == {**moved, "before": 30, "after": 66, "gain": 36, "label": "בדרך", "level_up": True}
|
| 30 |
assert readiness.change([("simulation", 90)], ("practice", 20))["gain"] == 0
|
| 31 |
+
|
| 32 |
+
|
| 33 |
+
def test_deepening_is_the_start_and_tests_pull_from_there():
|
| 34 |
+
assert readiness.compute([], 0.5)["number"] == 10 # half the skills mastered: 20% of 50
|
| 35 |
+
assert readiness.compute([], 1.0)["number"] == 20 and readiness.compute([], 1.0)["deepening"] == 20
|
| 36 |
+
assert readiness.compute([("simulation", 80)], 0.5)["number"] == 52 # 10, then 60% of the way to 80
|
| 37 |
+
assert readiness.compute([("simulation", 80)], 1.0)["can_add"]["simulation"] == 26 # at 56, a perfect simulation adds 60% of 44
|
| 38 |
+
more = [readiness.compute([("practice", 70)], share)["number"] for share in (0, 0.25, 0.5, 1)]
|
| 39 |
+
assert more == sorted(more) and more[0] < more[-1] # more mastery never lowers it
|
| 40 |
+
assert readiness.change([], ("practice", 80), 1.0) == {**readiness.change([], ("practice", 80), 1.0), "before": 20, "after": 38, "gain": 18}
|
mathly2/tests/test_realtests_api.py
CHANGED
|
@@ -364,17 +364,31 @@ def card_of(client, test_id, who=KID):
|
|
| 364 |
return client.get(f"/api/tests/{test_id}", headers=auth(who)).json()
|
| 365 |
|
| 366 |
|
| 367 |
-
def
|
| 368 |
client, store, lessons = world
|
| 369 |
test = set_test(client, lessons, days=14).json()
|
| 370 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 371 |
ready = card_of(client, test["id"])["readiness"]
|
| 372 |
-
assert ready["number"] ==
|
| 373 |
finished(store, test["id"], "exercise_test", f"{test['id']}:1", 80)
|
| 374 |
-
assert card_of(client, test["id"])["readiness"]["number"] ==
|
| 375 |
finished(store, test["id"], "simulation", str(test["id"]), 90)
|
| 376 |
ready = card_of(client, test["id"])["readiness"]
|
| 377 |
-
assert ready["number"] ==
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 378 |
|
| 379 |
|
| 380 |
def test_the_card_offers_three_practice_modes_simulations_and_the_trail_of_gains(world):
|
|
|
|
| 364 |
return client.get(f"/api/tests/{test_id}", headers=auth(who)).json()
|
| 365 |
|
| 366 |
|
| 367 |
+
def test_readiness_starts_from_deepening_and_tests_pull_it_up(world):
|
| 368 |
client, store, lessons = world
|
| 369 |
test = set_test(client, lessons, days=14).json()
|
| 370 |
+
skills = covered(client.app.state.bank, lessons[:2])
|
| 371 |
+
assert card_of(client, test["id"])["readiness"]["number"] == 0 # nothing mastered, no test
|
| 372 |
+
master(store, skills[:len(skills) // 2]) # half the material mastered in deepening
|
| 373 |
+
half = card_of(client, test["id"])["readiness"]
|
| 374 |
+
assert half["number"] == round(20 * (len(skills) // 2) / len(skills)) and half["deepening"] == half["number"]
|
| 375 |
+
master(store, skills) # all of it: 20% of the way from 0
|
| 376 |
ready = card_of(client, test["id"])["readiness"]
|
| 377 |
+
assert ready["number"] == 20 and ready["deepening"] == 20 and ready["label"] == "מתחילים" and ready["can_add"]["simulation"] == 48
|
| 378 |
finished(store, test["id"], "exercise_test", f"{test['id']}:1", 80)
|
| 379 |
+
assert card_of(client, test["id"])["readiness"]["number"] == 38 # 20 + 30% of (80 - 20)
|
| 380 |
finished(store, test["id"], "simulation", str(test["id"]), 90)
|
| 381 |
ready = card_of(client, test["id"])["readiness"]
|
| 382 |
+
assert ready["number"] == 69 and ready["practice_best"] == 80 and ready["simulation_best"] == 90 and ready["simulations"] == 1
|
| 383 |
+
|
| 384 |
+
|
| 385 |
+
def test_mastering_more_after_the_tests_still_raises_the_number(world):
|
| 386 |
+
client, store, lessons = world
|
| 387 |
+
test = set_test(client, lessons, days=14).json()
|
| 388 |
+
finished(store, test["id"], "exercise_test", f"{test['id']}:1", 60)
|
| 389 |
+
assert card_of(client, test["id"])["readiness"]["number"] == 18 # 30% of 60
|
| 390 |
+
master(store, covered(client.app.state.bank, lessons[:2]))
|
| 391 |
+
assert card_of(client, test["id"])["readiness"]["number"] == 32 # 20, then 20 + 30% of (60 - 20)
|
| 392 |
|
| 393 |
|
| 394 |
def test_the_card_offers_three_practice_modes_simulations_and_the_trail_of_gains(world):
|
mathly2/tests/test_writtentest_api.py
CHANGED
|
@@ -93,7 +93,7 @@ def test_after_handing_in_it_waits_for_the_check_and_the_ai_sets_the_grade():
|
|
| 93 |
assert got["checking"] is False and got["grade"] == attempts.percent(3, 2 * n) # 1 + ½ points of n
|
| 94 |
assert got["right"] == 1 and len(got["missed"]) == n - 1
|
| 95 |
assert got["missed"][0]["points"] == 0.5 and got["missed"][-1]["no_work"] is True
|
| 96 |
-
assert
|
| 97 |
|
| 98 |
|
| 99 |
def test_without_the_ai_it_waits_for_the_teacher_whose_points_are_final():
|
|
|
|
| 93 |
assert got["checking"] is False and got["grade"] == attempts.percent(3, 2 * n) # 1 + ½ points of n
|
| 94 |
assert got["right"] == 1 and len(got["missed"]) == n - 1
|
| 95 |
assert got["missed"][0]["points"] == 0.5 and got["missed"][-1]["no_work"] is True
|
| 96 |
+
assert all(set(m) == {"skill_id", "skill_name", "question", "picked", "points", "no_work", "mistake"} for m in got["missed"]) # never the right answer
|
| 97 |
|
| 98 |
|
| 99 |
def test_without_the_ai_it_waits_for_the_teacher_whose_points_are_final():
|
openspec/changes/readiness-counts-deepening/proposal.md
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Proposal
|
| 2 |
+
|
| 3 |
+
## Why
|
| 4 |
+
|
| 5 |
+
A student who masters the test's material in deepening is more ready, but the readiness number ignored it: only tests moved it. The teacher
|
| 6 |
+
wants deepening to count too, while tests stay the main way up.
|
| 7 |
+
|
| 8 |
+
## What Changes
|
| 9 |
+
|
| 10 |
+
- The share of the test's skills mastered sets where the number starts: it closes 20% of the gap from 0 (all of the material mastered starts it at 20).
|
| 11 |
+
- Tests then pull it up as before (a practice test 30% of the gap, a simulation 60%, a manual one 15%), and a lower grade never lowers it.
|
| 12 |
+
- Mastering more skills after the tests still raises the number.
|
| 13 |
+
- The test's page explains it, with what deepening gives now.
|
| 14 |
+
|
| 15 |
+
## Capabilities
|
| 16 |
+
|
| 17 |
+
### New Capabilities
|
| 18 |
+
|
| 19 |
+
### Modified Capabilities
|
| 20 |
+
- `readiness-meter`: deepening counts as the starting point.
|
| 21 |
+
|
| 22 |
+
## Impact
|
| 23 |
+
|
| 24 |
+
- `readiness.py` (the start and its share), `realtests.py` (the share of the test's skills mastered), `static/testcard.js` (the explanation). PRD RDY-2, RDY-4.
|
openspec/changes/readiness-counts-deepening/specs/readiness-meter/spec.md
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
## ADDED Requirements
|
| 2 |
+
|
| 3 |
+
### Requirement: Deepening counts
|
| 4 |
+
The readiness number SHALL start from deepening: the share of the test's skills the student has mastered, from 0 to 100, closes 20% of the gap from 0. The handed-in tests SHALL then move it from there, oldest first, as before, so mastering more skills raises the number whatever tests came after, and the test's page SHALL say how much deepening gives now.
|
| 5 |
+
|
| 6 |
+
#### Scenario: Half the material
|
| 7 |
+
- **WHEN** a student has mastered half of the test's skills and taken no test
|
| 8 |
+
- **THEN** the readiness number is 10
|
| 9 |
+
|
| 10 |
+
#### Scenario: Then a simulation
|
| 11 |
+
- **WHEN** that student hands in a simulation with 80
|
| 12 |
+
- **THEN** the number is 52 (10, then 60% of the way to 80)
|
openspec/changes/readiness-counts-deepening/tasks.md
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Tasks
|
| 2 |
+
|
| 3 |
+
- [x] 1.1 Start the number from the share of the test's skills mastered (`readiness.py`, `realtests.py`). Verify with `test_readiness.py` and `test_realtests_api.py`.
|
| 4 |
+
- [x] 1.2 Explain it on the test's page with what deepening gives now (`testcard.js`). Verify with the browser test of the explanation.
|
| 5 |
+
- [x] 1.3 Update the PRD (RDY-2, RDY-4). Verify that `openspec validate readiness-counts-deepening --strict` passes.
|