{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "434d0f98",
   "metadata": {},
   "source": [
    "# `newell-3det` — Newell's (1993) interior kinematic-wave reconstruction\n",
    "\n",
    "**What.** Newell's cumulative-count (N-curve) reformulation of the LWR model\n",
    "gives the interior traffic state, under a triangular fundamental diagram, as\n",
    "the MINIMUM of an uncongested trace (propagating forward at `+vf` from an\n",
    "upstream detector) and a congested trace (propagating backward at `-w` from a\n",
    "downstream detector) — no PDE solve. `ltm` already ships Newell's LOADING\n",
    "content (the min at the link ENDS, adr-016); `newell-3det` ships the unshipped\n",
    "INTERIOR content: given noisy/partial boundary detector curves, reconstruct\n",
    "the state at an interior point.\n",
    "\n",
    "**Why it is in the benchmark.** It is the benchmark's first traffic-STATE-\n",
    "ESTIMATION task (not an equilibrium principle): two reference estimators\n",
    "bracket a noisy detector card — the naive running-max baseline and an\n",
    "isotonic-regression-denoised alternative — scored against the harness-\n",
    "regenerated closed-form min. See the\n",
    "[model compendium](../../docs/MODELS.md) (Newell 1993) (P1).\n",
    "\n",
    "**Scope.** This notebook reconstructs the interior field on a clean spillback\n",
    "anchor (oracle row, RMSE 0 by construction) and on the seeded noisy\n",
    "discrimination card, certifying both and recomputing the interior\n",
    "Rankine-Hugoniot min-switch directly from the observed detector curves.\n",
    "\n",
    "**Canon.** `[newell1993simplified]`, [docs/REFERENCES.md](../../docs/REFERENCES.md) / [docs/references.bib](../../docs/references.bib)."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ab77d18",
   "metadata": {},
   "source": [
    "## How this notebook is graded\n",
    "\n",
    "**A notebook never claims a number it does not compute in that cell.** Every\n",
    "scored quantity below is recomputed live by the P1 `ThreeDetectorEvaluator`\n",
    "from the emitted interior field alone — the reference field is regenerated\n",
    "from the scenario's HASHED recipe (never a stored truth array), and the\n",
    "estimator's own provenance is never trusted\n",
    "([README](../../README.md), *Certified, not self-reported*)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "aa6ab5d1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:40.974166Z",
     "iopub.status.busy": "2026-07-21T13:49:40.973906Z",
     "iopub.status.idle": "2026-07-21T13:49:42.958668Z",
     "shell.execute_reply": "2026-07-21T13:49:42.957822Z"
    }
   },
   "outputs": [],
   "source": [
    "# Setup. `newell-3det` is a core model: a plain `pip install -e .` suffices —\n",
    "# no optional extra, so no guard cell. The inline backend is Agg-based\n",
    "# (headless CI renders into the notebook); NEVER matplotlib.use(\"Agg\")\n",
    "# in-kernel — it silently suppresses inline figure capture.\n",
    "%matplotlib inline\n",
    "import numpy as np\n",
    "\n",
    "from tabench import (\n",
    "    ThreeDetectorEvaluator,\n",
    "    newell_min,\n",
    "    newell_min_isotonic,\n",
    "    newell_spillback_scenario,\n",
    "    newell_noisy_scenario,\n",
    "    problem_from_scenario,\n",
    ")\n",
    "from tabench.dnl import interp_curve"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6bfeb852",
   "metadata": {},
   "source": [
    "## Anchor: a clean asymmetric spillback (oracle row)\n",
    "\n",
    "`vf=2, w=1, kappa=3, cap=2, L=4`, inflow 1.0 into a 0.5 meter — an interior\n",
    "Rankine-Hugoniot shock forms at the downstream end and travels upstream.\n",
    "`ThreeDetectorScenario` is frozen and content-hashed (P2); `noise='none'` makes\n",
    "this a validity/oracle row (never ranked), the anchor to certify the\n",
    "reconstruction pipeline itself is correct before trusting it on noisy data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fe80930a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:42.963446Z",
     "iopub.status.busy": "2026-07-21T13:49:42.963040Z",
     "iopub.status.idle": "2026-07-21T13:49:42.971800Z",
     "shell.execute_reply": "2026-07-21T13:49:42.971120Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "scenario      : newell-spillback\n",
      "content hash  : 48418c67ea3c12b6…\n",
      "vf=2.0, w=1.0, kappa=3.0, L=4.0, meter_cap=0.5\n",
      "interior query points : [1. 2. 3.]\n",
      "task          : reconstruct the interior cumulative field from boundary detectors\n"
     ]
    }
   ],
   "source": [
    "scenario = newell_spillback_scenario()\n",
    "print(f\"scenario      : {scenario.name}\")\n",
    "print(f\"content hash  : {scenario.content_hash()[:16]}…\")\n",
    "print(f\"vf={scenario.vf}, w={scenario.w}, kappa={scenario.kappa}, L={scenario.length}, \"\n",
    "      f\"meter_cap={scenario.meter_cap}\")\n",
    "print(f\"interior query points : {scenario.x_query}\")\n",
    "print(\"task          : reconstruct the interior cumulative field from boundary detectors\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "917b7f54",
   "metadata": {},
   "source": [
    "## Run\n",
    "\n",
    "No `Budget`/`RngBundle`/`Trace` — `problem_from_scenario` builds the MODEL-\n",
    "VISIBLE task (public physics + seeded boundary detectors, no truth recipe),\n",
    "and `newell_min` is a pure function of it, emitting a `ThreeDetectorField`\n",
    "(the P1-certifiable artifact)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "50b70a9c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:42.976080Z",
     "iopub.status.busy": "2026-07-21T13:49:42.975228Z",
     "iopub.status.idle": "2026-07-21T13:49:42.983294Z",
     "shell.execute_reply": "2026-07-21T13:49:42.982591Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "field shape (m, K+1) : (3, 25)\n",
      "interior curve at x=2 : [ 0.   0.   1.   2.   3.   4.   5.   6.   7.   8.   9.   9.5 10.  10.5\n",
      " 11.  11.5 12.  12.5 13.  13.5 14.  14.5 15.  15.5 16. ]\n"
     ]
    }
   ],
   "source": [
    "problem = problem_from_scenario(scenario)\n",
    "field = newell_min(problem)\n",
    "print(f\"field shape (m, K+1) : {field.field.shape}\")\n",
    "print(f\"interior curve at x=2 : {np.round(field.field[1], 3)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d70829c7",
   "metadata": {},
   "source": [
    "## Certify (P1) — regenerated from the hashed recipe, not a stored truth\n",
    "\n",
    "The harness reconstructs its OWN reference field from the scenario's content\n",
    "hash — it never reads a stored ground-truth array — and scores\n",
    "`interior_rmse` against it. On this clean anchor the reconstruction is exact."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a0593235",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:42.987299Z",
     "iopub.status.busy": "2026-07-21T13:49:42.986661Z",
     "iopub.status.idle": "2026-07-21T13:49:42.997651Z",
     "shell.execute_reply": "2026-07-21T13:49:42.996958Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "feasible       : 1\n",
      "interior_rmse  : 0.000e+00\n",
      "envelope_exact : 1\n",
      "  t=  9.5: N_up=8.500  N_dn=8.750  active=uncongested (+vf)\n",
      "  t= 10.0: N_up=9.000  N_dn=9.000  active=uncongested (+vf)\n",
      "  t= 10.5: N_up=9.500  N_dn=9.250  active=congested (-w)\n",
      "post-shock flow rate at x=2 : 0.5000  (meter cap: 0.5)\n"
     ]
    }
   ],
   "source": [
    "evaluator = ThreeDetectorEvaluator(scenario)\n",
    "metrics = evaluator.evaluate(field)\n",
    "print(f\"feasible       : {metrics['feasible']:.0f}\")\n",
    "print(f\"interior_rmse  : {metrics['interior_rmse']:.3e}\")\n",
    "print(f\"envelope_exact : {metrics['envelope_exact']:.0f}\")\n",
    "assert metrics[\"feasible\"] == 1.0\n",
    "assert metrics[\"interior_rmse\"] < 1e-9\n",
    "assert metrics[\"envelope_exact\"] == 1.0\n",
    "\n",
    "# DISTINCTIVE (Newell 1993): the interior state is the MIN of an uncongested\n",
    "# trace (+vf from upstream) and a congested trace (-w from downstream) —\n",
    "# recomputed here directly from the OBSERVED detector curves via the same\n",
    "# shift-and-min the production reconstruct_field uses, not a hand formula.\n",
    "up = problem.observation.up.mean(axis=0)\n",
    "dn = problem.observation.dn.mean(axis=0)\n",
    "vf, w, kappa, L = scenario.vf, scenario.w, scenario.kappa, scenario.length\n",
    "x = 2.0\n",
    "before, at, after = 9.5, 10.0, 10.5\n",
    "for t in (before, at, after):\n",
    "    n_up = interp_curve(up, t - x / vf, problem.dt)\n",
    "    n_dn = interp_curve(dn, t - (L - x) / w, problem.dt) + kappa * (L - x)\n",
    "    branch = \"uncongested (+vf)\" if n_up <= n_dn else \"congested (-w)\"\n",
    "    print(f\"  t={t:5.1f}: N_up={n_up:.3f}  N_dn={n_dn:.3f}  active={branch}\")\n",
    "# the min-switch at x=2 happens exactly at t=10 (both branches equal there).\n",
    "n_up_10 = interp_curve(up, at - x / vf, problem.dt)\n",
    "n_dn_10 = interp_curve(dn, at - (L - x) / w, problem.dt) + kappa * (L - x)\n",
    "assert np.isclose(n_up_10, n_dn_10, atol=1e-9)\n",
    "assert np.isclose(n_up_10, 9.0, atol=1e-9)\n",
    "# post-shock the interior flow rate (slope) settles at the meter cap q_B=0.5.\n",
    "post_shock_rate = (field.field[1, -1] - field.field[1, 12]) / (field.times[-1] - field.times[12])\n",
    "print(f\"post-shock flow rate at x=2 : {post_shock_rate:.4f}  (meter cap: {scenario.meter_cap})\")\n",
    "assert np.isclose(post_shock_rate, scenario.meter_cap, atol=1e-9)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3f6720fa",
   "metadata": {},
   "source": [
    "## The ranked task: naive running-max vs isotonic denoising\n",
    "\n",
    "`newell_noisy_scenario` applies the same spillback-family physics under a\n",
    "seeded Gaussian cumulative-reading level (`rankable=1`). The naive baseline\n",
    "makes the reading monotone with a running-max pass (biased high by every\n",
    "upward excursion); the isotonic baseline fits the L2-optimal nondecreasing\n",
    "curve first (averages the noise down) — recomputed and certified here, not\n",
    "quoted."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "f9a1a420",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:43.001590Z",
     "iopub.status.busy": "2026-07-21T13:49:43.001156Z",
     "iopub.status.idle": "2026-07-21T13:49:43.023834Z",
     "shell.execute_reply": "2026-07-21T13:49:43.023139Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "naive     interior_rmse : 1.0928\n",
      "isotonic  interior_rmse : 0.4484\n",
      "isotonic improvement    : 59.0%\n"
     ]
    }
   ],
   "source": [
    "noisy_scenario = newell_noisy_scenario()\n",
    "noisy_problem = problem_from_scenario(noisy_scenario)\n",
    "naive_field = newell_min(noisy_problem)\n",
    "iso_field = newell_min_isotonic(noisy_problem)\n",
    "\n",
    "noisy_evaluator = ThreeDetectorEvaluator(noisy_scenario)\n",
    "m_naive = noisy_evaluator.evaluate(naive_field)\n",
    "m_iso = noisy_evaluator.evaluate(iso_field)\n",
    "print(f\"naive     interior_rmse : {m_naive['interior_rmse']:.4f}\")\n",
    "print(f\"isotonic  interior_rmse : {m_iso['interior_rmse']:.4f}\")\n",
    "assert m_naive[\"feasible\"] == 1.0 and m_iso[\"feasible\"] == 1.0\n",
    "assert m_naive[\"rankable\"] == 1.0\n",
    "# The distinctive discrimination result: isotonic strictly beats naive.\n",
    "assert m_iso[\"interior_rmse\"] < m_naive[\"interior_rmse\"]\n",
    "print(f\"isotonic improvement    : {(1 - m_iso['interior_rmse'] / m_naive['interior_rmse']) * 100:.1f}%\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c72fd5f",
   "metadata": {},
   "source": [
    "## Visualize\n",
    "\n",
    "`tabench.viz` is a road-network flow visualizer; the newell-3det artifact is\n",
    "an interior cumulative-count FIELD over `(x, t)`, not a link-flow vector, so\n",
    "this notebook plots the certified interior curves directly (a house\n",
    "cumulative-curve plot, not `tabench.viz`)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "df2275c7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-21T13:49:43.027607Z",
     "iopub.status.busy": "2026-07-21T13:49:43.027048Z",
     "iopub.status.idle": "2026-07-21T13:49:43.265870Z",
     "shell.execute_reply": "2026-07-21T13:49:43.265304Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 950x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fig, axes = plt.subplots(1, 2, figsize=(9.5, 3.6))\n",
    "for i, x in enumerate(scenario.x_query):\n",
    "    axes[0].plot(field.times, field.field[i], label=f\"x={x:g}\")\n",
    "axes[0].axvline(10.0, color=\"0.6\", linestyle=\":\", label=\"shock at x=2\")\n",
    "axes[0].set_title(\"clean spillback (exact reconstruction)\")\n",
    "axes[0].set_xlabel(\"time\")\n",
    "axes[0].set_ylabel(\"cumulative count N(x,t)\")\n",
    "axes[0].legend(fontsize=7)\n",
    "\n",
    "axes[1].plot(naive_field.times, naive_field.field[1], label=f\"naive (rmse {m_naive['interior_rmse']:.2f})\")\n",
    "axes[1].plot(iso_field.times, iso_field.field[1], label=f\"isotonic (rmse {m_iso['interior_rmse']:.2f})\")\n",
    "axes[1].set_title(\"noisy card at x=2 (naive vs isotonic)\")\n",
    "axes[1].set_xlabel(\"time\")\n",
    "axes[1].legend(fontsize=7)\n",
    "fig.tight_layout()\n",
    "display(fig)\n",
    "plt.close(fig)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df6fc78e",
   "metadata": {},
   "source": [
    "## Takeaways & pointers\n",
    "\n",
    "- **Certified, not self-reported.** `ThreeDetectorEvaluator` regenerates the\n",
    "  reference field from the scenario's hashed recipe alone, never a stored\n",
    "  truth array; both estimators' RMSE came from that harness certificate.\n",
    "- **The min-switch is exact.** At the query point closest to the bottleneck,\n",
    "  the reconstructed shock crossing (t=10) and post-shock flow rate (0.5)\n",
    "  match the physics exactly, recomputed from the observed curves directly.\n",
    "- **Where next.** the loading twin that ships Newell's boundary content\n",
    "  [`ltm`](../05-dnl/02-ltm.ipynb); the lineage in the\n",
    "  [model compendium](../../docs/MODELS.md)."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.12"
  },
  "tabench": {
   "covers": [],
   "requires_extra": null,
   "track": "newell",
   "unit": "newell-3det"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}