{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "p5-00-ekrkgh",
   "metadata": {},
   "source": [
    "# Phase 5 — Strong Baselines & Fair Comparison\n",
    "## *Beyond WhAM* · CS 297 Final Paper\n",
    "\n",
    "---\n",
    "\n",
    "This notebook registers the **strong-baseline revision** of the paper, responding to\n",
    "the feedback that the original baselines were weak. We add 1-3 **strong baselines\n",
    "identified from the literature**, evaluate them through the *identical* frozen-embedding\n",
    "linear-probe protocol used in Phases 1-3, and add a **supervision-fairness analysis**\n",
    "of the headline individual-ID result.\n",
    "\n",
    "### Baselines and why each is justified (from literature)\n",
    "\n",
    "| Baseline | Tier | Literature basis |\n",
    "|---|---|---|\n",
    "| Raw ICI / Raw Mel | feature floor | linear model on raw features = lower bound *(logic)* |\n",
    "| **WhAM** (Paradise et al. 2025) | SOTA integrated | the model we build on; current SOTA representation for this data |\n",
    "| **AVES** (Hagiwara 2023) | strong, label-free | self-supervised animal-vocalization encoder; used as a baseline by WhAM itself |\n",
    "| **BirdNET** (Kahl et al. 2021) | strong, label-free | Tolkova et al. 2025 used BirdNET embeddings for **cetacean individual ID** on a small dataset like ours |\n",
    "\n",
    "### Two methodological points this notebook makes explicit\n",
    "1. **Tier by supervision.** DCCE-full learns with individual-ID labels (`L_id`); the\n",
    "   baselines never see ID labels. We therefore also train **DCCE without `L_id`** so our\n",
    "   architecture appears in the same label-free tier as the baselines (apples-to-apples).\n",
    "2. **No leakage.** For frozen baselines we report mean±std over 10 stratified splits.\n",
    "   For DCCE (which is *trained* on the seed-42 split) only the held-out seed-42 split is\n",
    "   leakage-free, so reshuffling would leak training codas into the probe test set. DCCE is\n",
    "   reported on the held-out split only.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-01-npiugo",
   "metadata": {},
   "source": [
    "## 1. Setup\n",
    "\n",
    "We reuse the shared evaluation harness (`scripts/lib_eval.py`), which reproduces the Phase 1 numbers exactly, so every baseline here is directly comparable to the existing tables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "p5-02-pchwkg",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:42:22.045943Z",
     "iopub.status.busy": "2026-07-01T19:42:22.045677Z",
     "iopub.status.idle": "2026-07-01T19:42:25.958350Z",
     "shell.execute_reply": "2026-07-01T19:42:25.957476Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "clean codas: 1383  |  ID-labeled: 762 (12 individuals)\n",
      "split: train 1106 / test 277  (seed 42, stratified by unit)\n"
     ]
    }
   ],
   "source": [
    "import os, sys\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "HERE = os.path.abspath(\".\")\n",
    "BASE = HERE if os.path.isdir(os.path.join(HERE, \"scripts\")) else os.path.dirname(HERE)\n",
    "sys.path.insert(0, os.path.join(BASE, \"scripts\"))\n",
    "DATA = os.path.join(BASE, \"datasets\")\n",
    "FIGS = os.path.join(BASE, \"figures\", \"phase5\"); os.makedirs(FIGS, exist_ok=True)\n",
    "\n",
    "from lib_eval import load_data, probe, probe_multiseed\n",
    "import run_baselines as rb\n",
    "import dcce_fairness as df_mod\n",
    "\n",
    "plt.rcParams.update({\"figure.dpi\": 120, \"font.size\": 11})\n",
    "D = load_data()\n",
    "X_mel_full = np.load(os.path.join(DATA, \"X_mel_full.npy\"))\n",
    "print(f\"clean codas: {len(D['df_clean'])}  |  ID-labeled: {len(D['df_id'])} \"\n",
    "      f\"({D['df_id']['individual_id'].nunique()} individuals)\")\n",
    "print(f\"split: train {len(D['train_idx'])} / test {len(D['test_idx'])}  \"\n",
    "      f\"(seed 42, stratified by unit)\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-03-sixauj",
   "metadata": {},
   "source": [
    "## 2. How the baseline embeddings were extracted\n",
    "\n",
    "All embeddings are frozen and aligned to `df_clean` order (n=1383). Extraction is kept\n",
    "in scripts (not inline) because two encoders need environments incompatible with this\n",
    "notebook's kernel:\n",
    "\n",
    "- **WhAM** — `datasets/wham_embeddings_all_layers.npy` `(1501, 20, 1280)`, indexed by\n",
    "  `coda_id - 1`. Extracted in Phase 2 from the VampNet coarse transformer (20 layers).\n",
    "- **AVES** — `datasets/emb_aves.npy` `(1383, 12, 768)`. Extracted by\n",
    "  `scripts/extract_ssl_embeddings.py aves` (main venv + `torchaudio`): the Earth Species\n",
    "  Project HuBERT-based AVES-base-bio checkpoint, mean-pooled per layer.\n",
    "- **BirdNET** — `datasets/emb_birdnet.npy` `(1383, 1024)`. Extracted by\n",
    "  `scripts/extract_birdnet.py` in a **separate Python 3.12 venv** (`.venv_birdnet`),\n",
    "  because BirdNET needs TensorFlow, which has no wheels for this project's Python 3.14\n",
    "  kernel. We read the 1024-d `GLOBAL_AVG_POOL/Mean` layer directly from the official\n",
    "  BirdNET V2.4 TFLite model (rebuilding the interpreter with\n",
    "  `experimental_preserve_all_tensors=True`, since current TFLite prunes intermediate\n",
    "  tensors and birdnetlib's own embedding path reads a null tensor).\n",
    "\n",
    "This notebook loads those `.npy` files and runs the probes here, so all *numbers* are\n",
    "reproduced from saved artifacts.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "p5-04-gnllau",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:42:25.960641Z",
     "iopub.status.busy": "2026-07-01T19:42:25.960478Z",
     "iopub.status.idle": "2026-07-01T19:42:25.997020Z",
     "shell.execute_reply": "2026-07-01T19:42:25.996486Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "WhAM (all layers)    OK (1501, 20, 1280)\n",
      "AVES                 OK (1383, 12, 768)\n",
      "BirdNET              OK (1383, 1024)\n"
     ]
    }
   ],
   "source": [
    "emb_files = {\n",
    "    \"WhAM (all layers)\": \"wham_embeddings_all_layers.npy\",\n",
    "    \"AVES\": \"emb_aves.npy\",\n",
    "    \"BirdNET\": \"emb_birdnet.npy\",\n",
    "}\n",
    "for name, fn in emb_files.items():\n",
    "    p = os.path.join(DATA, fn)\n",
    "    print(f\"{name:20s} {'OK ' + str(np.load(p, mmap_mode='r').shape) if os.path.exists(p) else 'MISSING'}\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-05-umxyum",
   "metadata": {},
   "source": [
    "## 3. Frozen baseline probing (floors + strong baselines)\n",
    "\n",
    "Each baseline is evaluated with the same logistic-regression linear probe on the same\n",
    "split. For multi-layer encoders (WhAM, AVES) we select the best layer **per task** —\n",
    "exactly as the paper does for WhAM (L10/L19). We report the single seed-42 split (matches\n",
    "existing tables) and mean±std over 10 stratified splits (error bars; valid because these\n",
    "encoders never saw our codas).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "p5-06-exztfr",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:42:25.998576Z",
     "iopub.status.busy": "2026-07-01T19:42:25.998471Z",
     "iopub.status.idle": "2026-07-01T19:44:07.893501Z",
     "shell.execute_reply": "2026-07-01T19:44:07.892174Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>tier</th>\n",
       "      <th>unit</th>\n",
       "      <th>type</th>\n",
       "      <th>id</th>\n",
       "      <th>unit±</th>\n",
       "      <th>type±</th>\n",
       "      <th>id±</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Raw ICI</td>\n",
       "      <td>floor</td>\n",
       "      <td>0.599</td>\n",
       "      <td>0.931</td>\n",
       "      <td>0.493</td>\n",
       "      <td>0.584</td>\n",
       "      <td>0.908</td>\n",
       "      <td>0.467</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Raw Mel</td>\n",
       "      <td>floor</td>\n",
       "      <td>0.740</td>\n",
       "      <td>0.097</td>\n",
       "      <td>0.280</td>\n",
       "      <td>0.765</td>\n",
       "      <td>0.135</td>\n",
       "      <td>0.287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>WhAM</td>\n",
       "      <td>label-free (SOTA)</td>\n",
       "      <td>0.895</td>\n",
       "      <td>0.261</td>\n",
       "      <td>0.426</td>\n",
       "      <td>0.909</td>\n",
       "      <td>0.227</td>\n",
       "      <td>0.394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AVES</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.905</td>\n",
       "      <td>0.290</td>\n",
       "      <td>0.418</td>\n",
       "      <td>0.916</td>\n",
       "      <td>0.237</td>\n",
       "      <td>0.402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BirdNET</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.863</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.322</td>\n",
       "      <td>0.897</td>\n",
       "      <td>0.231</td>\n",
       "      <td>0.402</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     model               tier   unit   type     id  unit±  type±    id±\n",
       "0  Raw ICI              floor  0.599  0.931  0.493  0.584  0.908  0.467\n",
       "1  Raw Mel              floor  0.740  0.097  0.280  0.765  0.135  0.287\n",
       "2     WhAM  label-free (SOTA)  0.895  0.261  0.426  0.909  0.227  0.394\n",
       "3     AVES         label-free  0.905  0.290  0.418  0.916  0.237  0.402\n",
       "4  BirdNET         label-free  0.863  0.272  0.322  0.897  0.231  0.402"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rows = []\n",
    "def add_row(name, tier, res):\n",
    "    rows.append(dict(model=name, tier=tier,\n",
    "                     **{f\"{t}_single\": res[t][\"macro_f1\"] for t in rb.TASKS},\n",
    "                     **{f\"{t}_mean\": res[t][\"f1_mean\"] for t in rb.TASKS},\n",
    "                     **{f\"{t}_std\": res[t][\"f1_std\"] for t in rb.TASKS},\n",
    "                     **{f\"{t}_layer\": res[t].get(\"layer\") for t in rb.TASKS}))\n",
    "\n",
    "# Floors\n",
    "for name, key in [(\"Raw ICI\", \"X_ici\")]:\n",
    "    add_row(name, \"floor\", rb.flat_eval(D[key], D))\n",
    "add_row(\"Raw Mel\", \"floor\", rb.flat_eval(np.load(os.path.join(DATA, \"X_mel_all.npy\")), D))\n",
    "\n",
    "# Strong baselines\n",
    "add_row(\"WhAM\", \"label-free (SOTA)\", rb.best_layer_eval(rb.load_wham_all(D), D)[0])\n",
    "add_row(\"AVES\", \"label-free\", rb.best_layer_eval(np.load(os.path.join(DATA, \"emb_aves.npy\")), D)[0])\n",
    "add_row(\"BirdNET\", \"label-free\", rb.flat_eval(np.load(os.path.join(DATA, \"emb_birdnet.npy\")), D))\n",
    "\n",
    "base_df = pd.DataFrame(rows)\n",
    "disp = base_df[[\"model\", \"tier\"] + [f\"{t}_single\" for t in rb.TASKS]\n",
    "               + [f\"{t}_mean\" for t in rb.TASKS]].round(3)\n",
    "disp.columns = [\"model\", \"tier\", \"unit\", \"type\", \"id\", \"unit±\", \"type±\", \"id±\"]\n",
    "disp\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-07-dnewra",
   "metadata": {},
   "source": [
    "## 4. DCCE fair-tier experiment (with vs. without `L_id`)\n",
    "\n",
    "We train the DCCE-`full` (cross-channel) variant twice using the verbatim Phase 3\n",
    "training loop (`scripts/dcce_fairness.py`):\n",
    "\n",
    "- **with `L_id`** (λ₂=0.5) — the paper's configuration; should reproduce individual-ID ≈ 0.834.\n",
    "- **without `L_id`** (λ₂=0) — removes individual-ID supervision so the representation is\n",
    "  label-free for ID, putting DCCE's *architecture* in the same tier as WhAM/AVES/BirdNET.\n",
    "\n",
    "Reported on the **held-out seed-42 split only** (leakage-free for a trained model).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "p5-08-heiqfe",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:44:07.898973Z",
     "iopub.status.busy": "2026-07-01T19:44:07.898823Z",
     "iopub.status.idle": "2026-07-01T19:45:36.292762Z",
     "shell.execute_reply": "2026-07-01T19:45:36.291893Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  trained (lambda2=0.5) in 46s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DCCE-full (with L_id)     unit=0.878  coda_type=0.578  individual_id=0.834\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  trained (lambda2=0.0) in 41s\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "DCCE-full (no L_id)       unit=0.883  coda_type=0.673  individual_id=0.571\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>model</th>\n",
       "      <th>tier</th>\n",
       "      <th>unit</th>\n",
       "      <th>coda_type</th>\n",
       "      <th>individual_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>DCCE-full (with L_id)</td>\n",
       "      <td>ID-supervised</td>\n",
       "      <td>0.878</td>\n",
       "      <td>0.578</td>\n",
       "      <td>0.834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DCCE-full (no L_id)</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.883</td>\n",
       "      <td>0.673</td>\n",
       "      <td>0.571</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   model           tier   unit  coda_type  individual_id\n",
       "0  DCCE-full (with L_id)  ID-supervised  0.878      0.578          0.834\n",
       "1    DCCE-full (no L_id)     label-free  0.883      0.673          0.571"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dcce_rows = []\n",
    "for tag, lam2, tier in [(\"DCCE-full (with L_id)\", 0.5, \"ID-supervised\"),\n",
    "                        (\"DCCE-full (no L_id)\", 0.0, \"label-free\")]:\n",
    "    model = df_mod.train_full(D, X_mel_full, lambda2=lam2)\n",
    "    Z = df_mod.extract_z(model, D[\"X_ici\"], X_mel_full, np.arange(len(D[\"df_clean\"])))\n",
    "    r = probe(Z, D)   # single-split, seed 42, held-out test\n",
    "    dcce_rows.append(dict(model=tag, tier=tier,\n",
    "                          **{t: r[t][\"macro_f1\"] for t in rb.TASKS}))\n",
    "    print(f\"{tag:24s}  \" + \"  \".join(f\"{t}={r[t]['macro_f1']:.3f}\" for t in rb.TASKS))\n",
    "\n",
    "dcce_df = pd.DataFrame(dcce_rows)\n",
    "dcce_df.round(3)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-09-bpqrif",
   "metadata": {},
   "source": [
    "## 5. Consolidated tiered table (single-split seed-42)\n",
    "\n",
    "The leakage-free head-to-head: all models on the same held-out split, grouped by tier.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "p5-10-otvwpe",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:45:36.295416Z",
     "iopub.status.busy": "2026-07-01T19:45:36.295250Z",
     "iopub.status.idle": "2026-07-01T19:45:36.306351Z",
     "shell.execute_reply": "2026-07-01T19:45:36.305769Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Model</th>\n",
       "      <th>Tier</th>\n",
       "      <th>Unit</th>\n",
       "      <th>Coda Type</th>\n",
       "      <th>Individual ID</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Raw ICI</td>\n",
       "      <td>floor</td>\n",
       "      <td>0.599</td>\n",
       "      <td>0.931</td>\n",
       "      <td>0.493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Raw Mel</td>\n",
       "      <td>floor</td>\n",
       "      <td>0.740</td>\n",
       "      <td>0.097</td>\n",
       "      <td>0.280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>WhAM</td>\n",
       "      <td>label-free (SOTA)</td>\n",
       "      <td>0.895</td>\n",
       "      <td>0.261</td>\n",
       "      <td>0.426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AVES</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.905</td>\n",
       "      <td>0.290</td>\n",
       "      <td>0.418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BirdNET</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.863</td>\n",
       "      <td>0.272</td>\n",
       "      <td>0.322</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>DCCE-full (no L_id)</td>\n",
       "      <td>label-free</td>\n",
       "      <td>0.883</td>\n",
       "      <td>0.673</td>\n",
       "      <td>0.571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>DCCE-full (with L_id)</td>\n",
       "      <td>ID-supervised</td>\n",
       "      <td>0.878</td>\n",
       "      <td>0.578</td>\n",
       "      <td>0.834</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   Model               Tier   Unit  Coda Type  Individual ID\n",
       "0                Raw ICI              floor  0.599      0.931          0.493\n",
       "1                Raw Mel              floor  0.740      0.097          0.280\n",
       "2                   WhAM  label-free (SOTA)  0.895      0.261          0.426\n",
       "3                   AVES         label-free  0.905      0.290          0.418\n",
       "4                BirdNET         label-free  0.863      0.272          0.322\n",
       "5    DCCE-full (no L_id)         label-free  0.883      0.673          0.571\n",
       "6  DCCE-full (with L_id)      ID-supervised  0.878      0.578          0.834"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "consolidated = []\n",
    "for _, r in base_df.iterrows():\n",
    "    consolidated.append(dict(Model=r[\"model\"], Tier=r[\"tier\"],\n",
    "                             Unit=r[\"unit_single\"], **{\"Coda Type\": r[\"coda_type_single\"]},\n",
    "                             **{\"Individual ID\": r[\"individual_id_single\"]}))\n",
    "for _, r in dcce_df.iterrows():\n",
    "    consolidated.append(dict(Model=r[\"model\"], Tier=r[\"tier\"], Unit=r[\"unit\"],\n",
    "                             **{\"Coda Type\": r[\"coda_type\"]}, **{\"Individual ID\": r[\"individual_id\"]}))\n",
    "tier_order = {\"floor\": 0, \"label-free (SOTA)\": 1, \"label-free\": 1, \"ID-supervised\": 2}\n",
    "ctab = pd.DataFrame(consolidated)\n",
    "ctab = ctab.sort_values(by=\"Tier\", key=lambda s: s.map(tier_order)).reset_index(drop=True)\n",
    "ctab.round(3)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-11-ugruky",
   "metadata": {},
   "source": [
    "## 6. The two linguistic channels (primary): coda type + social unit\n",
    "\n",
    "Understanding sperm whale *communication* is primarily about the structure of the codas —\n",
    "the **coda-type vocabulary** (rhythm channel) and the **social-unit dialect** (identity\n",
    "channel) — rather than identifying *which individual* is speaking. We therefore treat unit\n",
    "and coda type as the primary tasks.\n",
    "\n",
    "The key question for a language-oriented representation is not \"does it top one task?\" but\n",
    "**\"does it capture *both* channels?\"** A representation that nails the dialect but is blind\n",
    "to the rhythmic vocabulary (or vice versa) has only half the language.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "p5-12-ipopzy",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:45:36.307988Z",
     "iopub.status.busy": "2026-07-01T19:45:36.307830Z",
     "iopub.status.idle": "2026-07-01T19:45:36.315939Z",
     "shell.execute_reply": "2026-07-01T19:45:36.315317Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best on Unit      : AVES (0.905)\n",
      "Best on Coda Type : Raw ICI (0.931)\n",
      "\n",
      "Spanning BOTH channels (higher min = does not collapse on either):\n",
      "                Model  Unit  Coda Type  mean(U,T)  min(U,T)\n",
      "  DCCE-full (no L_id) 0.883      0.673      0.778     0.673\n",
      "              Raw ICI 0.599      0.931      0.765     0.599\n",
      "DCCE-full (with L_id) 0.878      0.578      0.728     0.578\n",
      "                 AVES 0.905      0.290      0.597     0.290\n",
      "              BirdNET 0.863      0.272      0.567     0.272\n",
      "                 WhAM 0.895      0.261      0.578     0.261\n",
      "              Raw Mel 0.740      0.097      0.418     0.097\n"
     ]
    }
   ],
   "source": [
    "color_map = {\"floor\": \"#BBBBBB\", \"label-free (SOTA)\": \"#4C72B0\",\n",
    "             \"label-free\": \"#55A868\", \"ID-supervised\": \"#C44E52\"}\n",
    "\n",
    "# Per-task winners (honest: DCCE does not top either task alone)\n",
    "for task in [\"Unit\", \"Coda Type\"]:\n",
    "    top = ctab.sort_values(task, ascending=False).iloc[0]\n",
    "    print(f\"Best on {task:10s}: {top['Model']} ({top[task]:.3f})\")\n",
    "\n",
    "# Balance across BOTH channels: mean and worst-case (min)\n",
    "lang = ctab.copy()\n",
    "lang[\"mean(U,T)\"] = lang[[\"Unit\", \"Coda Type\"]].mean(axis=1)\n",
    "lang[\"min(U,T)\"]  = lang[[\"Unit\", \"Coda Type\"]].min(axis=1)\n",
    "lang = lang.sort_values(\"min(U,T)\", ascending=False).reset_index(drop=True)\n",
    "print(\"\\nSpanning BOTH channels (higher min = does not collapse on either):\")\n",
    "print(lang[[\"Model\", \"Unit\", \"Coda Type\", \"mean(U,T)\", \"min(U,T)\"]].round(3).to_string(index=False))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-13-necflt",
   "metadata": {},
   "source": [
    "### Spanning both channels — the key figure\n",
    "\n",
    "Plotting coda-type F1 (x) against social-unit F1 (y) makes the structure obvious: the\n",
    "audio encoders cluster **top-left** (dialect yes, vocabulary no), raw ICI sits **bottom-right**\n",
    "(vocabulary yes, dialect no), and only DCCE reaches the **upper-right** region where *both*\n",
    "channels are captured.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "p5-14-fldbut",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:45:36.319760Z",
     "iopub.status.busy": "2026-07-01T19:45:36.319606Z",
     "iopub.status.idle": "2026-07-01T19:45:36.632491Z",
     "shell.execute_reply": "2026-07-01T19:45:36.631936Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 936x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(7.8, 6))\n",
    "for _, r in ctab.iterrows():\n",
    "    ax.scatter(r[\"Coda Type\"], r[\"Unit\"], s=110, color=color_map[r[\"Tier\"]],\n",
    "               edgecolor=\"black\", zorder=3)\n",
    "    ax.annotate(r[\"Model\"], (r[\"Coda Type\"], r[\"Unit\"]), fontsize=8.5,\n",
    "                xytext=(6, 4), textcoords=\"offset points\")\n",
    "ax.set_xlabel(\"Coda Type macro-F1  (rhythm channel — the 'vocabulary')\")\n",
    "ax.set_ylabel(\"Social Unit macro-F1  (dialect / identity channel)\")\n",
    "ax.set_title(\"Does the representation span BOTH linguistic channels?\\n\"\n",
    "             \"(upper-right = captures rhythm AND dialect)\")\n",
    "ax.set_xlim(0, 1); ax.set_ylim(0.5, 1.0)\n",
    "ax.axhspan(0.82, 1.0, xmin=0.5, alpha=0.06, color=\"green\")\n",
    "ax.grid(alpha=0.2)\n",
    "from matplotlib.patches import Patch\n",
    "ax.legend(handles=[Patch(color=c, label=l) for l, c in color_map.items()],\n",
    "          fontsize=8, loc=\"lower left\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(os.path.join(FIGS, \"fig_two_channels.png\"), dpi=130, bbox_inches=\"tight\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-15-ngdqxn",
   "metadata": {},
   "source": [
    "## 7. Individual ID (secondary evidence)\n",
    "\n",
    "Individual ID is a \"who is speaking\" biometric rather than a property of the *language*,\n",
    "so we report it as **secondary** evidence: it shows whether a representation additionally\n",
    "captures fine speaker structure that the others miss. (Tiered by supervision, since\n",
    "DCCE-full sees ID labels and the baselines do not.)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "p5-16-efihnr",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-01T19:45:36.633884Z",
     "iopub.status.busy": "2026-07-01T19:45:36.633774Z",
     "iopub.status.idle": "2026-07-01T19:45:36.798282Z",
     "shell.execute_reply": "2026-07-01T19:45:36.797748Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                Model              Tier  Individual ID\n",
      "              Raw Mel             floor          0.280\n",
      "              BirdNET        label-free          0.322\n",
      "                 AVES        label-free          0.418\n",
      "                 WhAM label-free (SOTA)          0.426\n",
      "              Raw ICI             floor          0.493\n",
      "  DCCE-full (no L_id)        label-free          0.571\n",
      "DCCE-full (with L_id)     ID-supervised          0.834\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1080x540 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "id_tab = ctab[[\"Model\", \"Tier\", \"Individual ID\"]].sort_values(\"Individual ID\").reset_index(drop=True)\n",
    "print(id_tab.round(3).to_string(index=False))\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(9, 4.5))\n",
    "bars = ax.bar(id_tab[\"Model\"], id_tab[\"Individual ID\"],\n",
    "              color=[color_map[t] for t in id_tab[\"Tier\"]], edgecolor=\"black\", linewidth=0.5)\n",
    "ax.set_ylabel(\"Individual ID macro-F1\"); ax.set_ylim(0, 0.9)\n",
    "ax.set_title(\"Individual ID (secondary): tiered by supervision, held-out seed-42 split\")\n",
    "ax.tick_params(axis=\"x\", rotation=30)\n",
    "for b, v in zip(bars, id_tab[\"Individual ID\"]):\n",
    "    ax.text(b.get_x() + b.get_width()/2, v + 0.01, f\"{v:.3f}\", ha=\"center\", fontsize=9)\n",
    "from matplotlib.patches import Patch\n",
    "ax.legend(handles=[Patch(color=c, label=l) for l, c in color_map.items()], fontsize=8, loc=\"upper left\")\n",
    "plt.tight_layout()\n",
    "plt.savefig(os.path.join(FIGS, \"fig_individual_id_tiers.png\"), dpi=130, bbox_inches=\"tight\")\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "p5-17-tievga",
   "metadata": {},
   "source": [
    "## 8. Interpretation\n",
    "\n",
    "**Primary (the two linguistic channels).**\n",
    "- **No single baseline captures both channels.** Audio encoders (WhAM 0.26, AVES 0.29,\n",
    "  BirdNET 0.27 on coda type) are strong on the social-unit dialect (~0.90) but nearly\n",
    "  blind to the rhythmic vocabulary; raw ICI is the mirror image (coda type 0.931, unit\n",
    "  0.599). Each has only half the language.\n",
    "- **DCCE is the only representation strong on both** — social unit 0.883 *and* coda type\n",
    "  0.673 (DCCE without `L_id`). Measured by worst-case-across-channels, DCCE's `min(U,T)`\n",
    "  = 0.673 versus the next-best 0.599 (raw ICI) and ~0.27 for the audio encoders.\n",
    "- **Honest caveat:** DCCE does **not** top either task individually — AVES/WhAM edge it on\n",
    "  social unit (~0.90 vs 0.883) and raw ICI dominates coda type (0.931 vs 0.673). The\n",
    "  contribution is being the best *joint* encoder of both channels, not the best per task.\n",
    "- **Why this happens:** the two channels live in different feature spaces — rhythm/coda\n",
    "  type in the symbolic ICI timing, social unit in the spectral audio texture — so only a\n",
    "  dual-channel design spans both. This is direct evidence for the two-channel premise.\n",
    "\n",
    "**Secondary (individual ID).** With no ID supervision, DCCE's architecture still leads the\n",
    "label-free tier (0.571 vs WhAM 0.426, AVES 0.418, BirdNET 0.322), evidence it additionally\n",
    "captures speaker structure the other encoders miss; ID supervision raises this to 0.834.\n",
    "\n",
    "*Numbers here reproduce the consolidated analysis in `reports/fair_comparison.md`.*\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.14.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}