{ "cells": [ { "cell_type": "markdown", "id": "a086f802", "metadata": {}, "source": [ "# 03B — Model Evolution: V1 to V2\n", "\n", "**Mengapa enam cluster disederhanakan menjadi tiga tipologi**\n", "\n", "Notebook ini adalah POC portfolio yang dapat dijalankan ulang dari artefak lokal Pasuruan Lens. Kode produksi tetap berada di `python/scripts/` dan `python/src/pasuruan365/`; notebook berfungsi sebagai narasi analitik yang ringkas, transparan, dan mudah dipresentasikan." ] }, { "cell_type": "code", "execution_count": 1, "id": "1d189422", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:25.783262Z", "iopub.status.busy": "2026-09-04T06:27:25.783108Z", "iopub.status.idle": "2026-09-04T06:27:27.300677Z", "shell.execute_reply": "2026-09-04T06:27:27.299636Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Project root: PASURUAN365\n", "Python: 3.12.6\n" ] } ], "source": [ "from pathlib import Path\n", "import json\n", "import sys\n", "\n", "import matplotlib.pyplot as plt\n", "import numpy as np\n", "import pandas as pd\n", "import seaborn as sns\n", "\n", "\n", "def find_project_root(start: Path | None = None) -> Path:\n", " start = (start or Path.cwd()).resolve()\n", " for candidate in (start, *start.parents):\n", " if (candidate / \"package.json\").exists() and (candidate / \"data\").exists():\n", " return candidate\n", " raise RuntimeError(\"Root Pasuruan Lens tidak ditemukan. Jalankan notebook dari repository ini.\")\n", "\n", "\n", "ROOT = find_project_root()\n", "sys.path.insert(0, str(ROOT / \"python\" / \"src\"))\n", "\n", "pd.set_option(\"display.max_columns\", 100)\n", "pd.set_option(\"display.max_colwidth\", 100)\n", "sns.set_theme(style=\"whitegrid\", context=\"notebook\")\n", "COLORS = {\"A\": \"#2563eb\", \"B\": \"#f59e0b\", \"C\": \"#10b981\"}\n", "\n", "print(f\"Project root: {ROOT}\")\n", "print(f\"Python: {sys.version.split()[0]}\")" ] }, { "cell_type": "markdown", "id": "7b696192", "metadata": {}, "source": [ "## Executive takeaway\n", "\n", "V1 (`k=6`) adalah baseline yang valid dan cukup stabil ketika pipeline yang sama diulang. Namun, robustness review menemukan dua masalah: assignment sangat berubah ketika scaler diganti dan fitur poliklinik berbentuk laju `log1p` sangat zero-inflated. V2 tidak dipilih karena menang di semua metrik; V2 dipilih karena menjadi konfigurasi paling kuat **di antara kandidat yang lolos gate semantik, cross-scaler, dan pipeline-consensus**.\n", "\n", "```text\n", "V1: 8 fitur → StandardScaler → PCA 6 → K-Means k=6\n", " ↓ robustness review\n", "5 feature sets × 2 scalers × PCA/original × 2 algorithms × k=3…8\n", " ↓ semantic + stability gates\n", "V2: 8 fitur → StandardScaler → PCA 7 → K-Means k=3\n", "```" ] }, { "cell_type": "code", "execution_count": 2, "id": "7fa374da", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:27.304509Z", "iopub.status.busy": "2026-09-04T06:27:27.304072Z", "iopub.status.idle": "2026-09-04T06:27:27.686996Z", "shell.execute_reply": "2026-09-04T06:27:27.686280Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
nilai
unit analisis365
seluruh eksperimen504
finalis dengan evaluasi mendalam25
feature sets5
rentang k3–8
algoritmaAGGLOMERATIVE_WARD, KMEANS
\n", "
" ], "text/plain": [ " nilai\n", "unit analisis 365\n", "seluruh eksperimen 504\n", "finalis dengan evaluasi mendalam 25\n", "feature sets 5\n", "rentang k 3–8\n", "algoritma AGGLOMERATIVE_WARD, KMEANS" ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "from sklearn.cluster import KMeans\n", "from sklearn.decomposition import PCA\n", "from sklearn.metrics import adjusted_rand_score, normalized_mutual_info_score, silhouette_score\n", "from sklearn.preprocessing import StandardScaler\n", "\n", "eda = pd.read_csv(ROOT / \"data/analytics/village_eda_features.csv\", dtype={\"village_id\": str})\n", "v1 = pd.read_csv(ROOT / \"data/modeling/cluster_assignments.csv\", dtype={\"village_id\": str})\n", "v2 = pd.read_csv(ROOT / \"data/modeling/cluster_assignments_v2.csv\", dtype={\"village_id\": str})\n", "v1_config = json.loads((ROOT / \"data/modeling/selected_model_config.json\").read_text())\n", "v2_config = json.loads((ROOT / \"data/modeling/selected_model_config_v2.json\").read_text())\n", "experiments = pd.read_csv(ROOT / \"reports/task05b/candidate_experiments.csv\")\n", "finalists = pd.read_csv(ROOT / \"reports/task05b/final_candidate_models.csv\")\n", "poly_review = pd.read_csv(ROOT / \"reports/task05b/polyclinic_representation_comparison.csv\")\n", "feature_ablation = pd.read_csv(ROOT / \"reports/task05b/feature_ablation.csv\")\n", "\n", "pd.Series({\n", " \"unit analisis\": eda.village_id.nunique(),\n", " \"seluruh eksperimen\": len(experiments),\n", " \"finalis dengan evaluasi mendalam\": len(finalists),\n", " \"feature sets\": experiments.feature_set.nunique(),\n", " \"rentang k\": f\"{experiments.k.min()}–{experiments.k.max()}\",\n", " \"algoritma\": \", \".join(sorted(experiments.algorithm.unique())),\n", "}).to_frame(\"nilai\")" ] }, { "cell_type": "markdown", "id": "f9d6bb8e", "metadata": {}, "source": [ "## 1 — Reproduksi V1 dan V2" ] }, { "cell_type": "code", "execution_count": 3, "id": "f2bea742", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:27.689567Z", "iopub.status.busy": "2026-09-04T06:27:27.689337Z", "iopub.status.idle": "2026-09-04T06:27:27.890338Z", "shell.execute_reply": "2026-09-04T06:27:27.888895Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
modelkARI_vs_frozen_assignmentreproduced_silhouette
0V161.00.222292
1V231.00.175881
\n", "
" ], "text/plain": [ " model k ARI_vs_frozen_assignment reproduced_silhouette\n", "0 V1 6 1.0 0.222292\n", "1 V2 3 1.0 0.175881" ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "def reproduce(config, frame):\n", " values = frame.copy()\n", " if \"has_polyclinic\" in config[\"features\"]:\n", " values[\"has_polyclinic\"] = values[\"polyclinic_count\"].gt(0).astype(int)\n", " X = values[config[\"features\"]].astype(float)\n", " scaled = StandardScaler().fit_transform(X)\n", " n_components = config.get(\"number_of_pca_components\", config.get(\"pca_components\"))\n", " projected = PCA(n_components=n_components, random_state=config[\"random_state\"]).fit_transform(scaled)\n", " labels = KMeans(\n", " n_clusters=config[\"k\"], random_state=config[\"random_state\"], n_init=20\n", " ).fit_predict(projected)\n", " return projected, labels\n", "\n", "\n", "v1_space, v1_reproduced = reproduce(v1_config, eda)\n", "v2_space, v2_reproduced = reproduce(v2_config, eda)\n", "\n", "reproduction = pd.DataFrame({\n", " \"model\": [\"V1\", \"V2\"],\n", " \"k\": [v1_config[\"k\"], v2_config[\"k\"]],\n", " \"ARI_vs_frozen_assignment\": [\n", " adjusted_rand_score(v1.cluster_id, v1_reproduced),\n", " adjusted_rand_score(v2.cluster_id, v2_reproduced),\n", " ],\n", " \"reproduced_silhouette\": [\n", " silhouette_score(v1_space, v1_reproduced),\n", " silhouette_score(v2_space, v2_reproduced),\n", " ],\n", "})\n", "reproduction.round(6)" ] }, { "cell_type": "code", "execution_count": 4, "id": "be166715", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:27.894360Z", "iopub.status.busy": "2026-09-04T06:27:27.893939Z", "iopub.status.idle": "2026-09-04T06:27:27.899844Z", "shell.execute_reply": "2026-09-04T06:27:27.899000Z" } }, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "✓ V1 dan V2 dapat direproduksi dari feature layer dengan ARI permutation-invariant ≈ 1.\n" ] } ], "source": [ "assert reproduction.ARI_vs_frozen_assignment.min() > .999\n", "print(\"✓ V1 dan V2 dapat direproduksi dari feature layer dengan ARI permutation-invariant ≈ 1.\")" ] }, { "cell_type": "markdown", "id": "b44b6f4e", "metadata": {}, "source": [ "ARI dipakai karena nomor cluster bersifat nominal: cluster `1` pada dua hasil fit tidak harus memiliki nomor yang sama untuk mewakili partisi yang sama." ] }, { "cell_type": "markdown", "id": "f5792198", "metadata": {}, "source": [ "## 2 — Apa yang memicu review V1?" ] }, { "cell_type": "code", "execution_count": 5, "id": "4e3b1308", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:27.904060Z", "iopub.status.busy": "2026-09-04T06:27:27.903656Z", "iopub.status.idle": "2026-09-04T06:27:27.922154Z", "shell.execute_reply": "2026-09-04T06:27:27.921058Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
nilai
V1 silhouette0.222
V1 bootstrap stability ARI0.939
V1 Standard-vs-Robust ARI0.270
Polyclinic zero percentage86.849
Polyclinic IQR0.000
\n", "
" ], "text/plain": [ " nilai\n", "V1 silhouette 0.222\n", "V1 bootstrap stability ARI 0.939\n", "V1 Standard-vs-Robust ARI 0.270\n", "Polyclinic zero percentage 86.849\n", "Polyclinic IQR 0.000" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "v1_candidate_id = \"T05B_A_CURRENT_8_KM_STANDARD_PCA6_K6\"\n", "v2_candidate_id = \"T05B_C_BINARY_POLYCLINIC_8_KM_STANDARD_PCA7_K3\"\n", "v1_row = finalists.loc[finalists.model_id.eq(v1_candidate_id)].iloc[0]\n", "v2_row = finalists.loc[finalists.model_id.eq(v2_candidate_id)].iloc[0]\n", "\n", "trigger_summary = pd.Series({\n", " \"V1 silhouette\": v1_config[\"metrics\"][\"silhouette\"],\n", " \"V1 bootstrap stability ARI\": v1_config[\"metrics\"][\"bootstrap_mean_ari\"],\n", " \"V1 Standard-vs-Robust ARI\": v1_row[\"scaler_robustness\"],\n", " \"Polyclinic zero percentage\": eda[\"polyclinic_count\"].eq(0).mean() * 100,\n", " \"Polyclinic IQR\": eda[\"polyclinic_count\"].quantile(.75) - eda[\"polyclinic_count\"].quantile(.25),\n", "}).round(3)\n", "trigger_summary.to_frame(\"nilai\")" ] }, { "cell_type": "code", "execution_count": 6, "id": "bb093cc3", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:27.925920Z", "iopub.status.busy": "2026-09-04T06:27:27.925529Z", "iopub.status.idle": "2026-09-04T06:27:28.297780Z", "shell.execute_reply": "2026-09-04T06:27:28.297358Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
removed_featuredomainari_vs_v1fragility_score_1_minus_ari
7cellular_operator_countCONNECTIVITY0.8050.195
6log1p_bts_per_10000_populationCONNECTIVITY0.7950.205
5log1p_population_densityGEOGRAPHY0.7910.209
4population_share_of_districtDEMOGRAPHY0.7810.219
3log1p_polyclinic_per_10000_populationHEALTHCARE0.5830.417
2sex_ratioDEMOGRAPHY0.5820.418
1has_banking_facilityECONOMY0.5750.425
0any_documented_disaster_event_2024DISASTER0.5040.496
\n", "
" ], "text/plain": [ " removed_feature domain ari_vs_v1 \\\n", "7 cellular_operator_count CONNECTIVITY 0.805 \n", "6 log1p_bts_per_10000_population CONNECTIVITY 0.795 \n", "5 log1p_population_density GEOGRAPHY 0.791 \n", "4 population_share_of_district DEMOGRAPHY 0.781 \n", "3 log1p_polyclinic_per_10000_population HEALTHCARE 0.583 \n", "2 sex_ratio DEMOGRAPHY 0.582 \n", "1 has_banking_facility ECONOMY 0.575 \n", "0 any_documented_disaster_event_2024 DISASTER 0.504 \n", "\n", " fragility_score_1_minus_ari \n", "7 0.195 \n", "6 0.205 \n", "5 0.209 \n", "4 0.219 \n", "3 0.417 \n", "2 0.418 \n", "1 0.425 \n", "0 0.496 " ] }, "execution_count": 6, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "ablation_plot = feature_ablation.sort_values(\"fragility_score_1_minus_ari\")\n", "fig, ax = plt.subplots(figsize=(9, 5))\n", "bars = ax.barh(\n", " ablation_plot.removed_feature.str.replace(\"log1p_\", \"\", regex=False),\n", " ablation_plot.fragility_score_1_minus_ari,\n", " color=\"#f59e0b\",\n", ")\n", "ax.set(\n", " title=\"V1 berubah material ketika satu fitur dihapus\",\n", " xlabel=\"Fragility = 1 − ARI (lebih tinggi = lebih sensitif)\",\n", " ylabel=\"Fitur yang dihapus\",\n", " xlim=(0, .55),\n", ")\n", "ax.bar_label(bars, fmt=\"%.2f\", padding=3, fontsize=9)\n", "ax.spines[[\"top\", \"right\"]].set_visible(False)\n", "plt.tight_layout()\n", "\n", "ablation_plot[[\"removed_feature\", \"domain\", \"ari_vs_v1\", \"fragility_score_1_minus_ari\"]].round(3)" ] }, { "cell_type": "markdown", "id": "81a1e00d", "metadata": {}, "source": [ "V1 memiliki separation dan repeatability yang baik, tetapi repeatability **di dalam satu spesifikasi** tidak sama dengan robustness terhadap pilihan preprocessing atau feature representation yang sama-sama masuk akal." ] }, { "cell_type": "markdown", "id": "a668adca", "metadata": {}, "source": [ "## 3 — Audit representasi poliklinik" ] }, { "cell_type": "code", "execution_count": 7, "id": "61fc4bb1", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.299732Z", "iopub.status.busy": "2026-09-04T06:27:28.299530Z", "iopub.status.idle": "2026-09-04T06:27:28.313566Z", "shell.execute_reply": "2026-09-04T06:27:28.313190Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
representationcross_scaler_ARImean_silhouettemin_cluster_size
4RAW_COUNT0.3460.20023
2EXCLUDED0.3090.19229
1CURRENT_LOG1P_RATE0.2700.19428
0BINARY_PRESENCE0.2650.19627
3LOG1P_COUNT0.2500.19425
\n", "
" ], "text/plain": [ " representation cross_scaler_ARI mean_silhouette min_cluster_size\n", "4 RAW_COUNT 0.346 0.200 23\n", "2 EXCLUDED 0.309 0.192 29\n", "1 CURRENT_LOG1P_RATE 0.270 0.194 28\n", "0 BINARY_PRESENCE 0.265 0.196 27\n", "3 LOG1P_COUNT 0.250 0.194 25" ] }, "execution_count": 7, "metadata": {}, "output_type": "execute_result" } ], "source": [ "poly_table = (\n", " poly_review.groupby(\"representation\", as_index=False)\n", " .agg(\n", " cross_scaler_ARI=(\"scaler_agreement_ari\", \"first\"),\n", " mean_silhouette=(\"silhouette\", \"mean\"),\n", " min_cluster_size=(\"min_cluster_size\", \"min\"),\n", " )\n", " .sort_values(\"cross_scaler_ARI\", ascending=False)\n", ")\n", "poly_table.round(3)" ] }, { "cell_type": "code", "execution_count": 8, "id": "2199d3d0", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.315386Z", "iopub.status.busy": "2026-09-04T06:27:28.315195Z", "iopub.status.idle": "2026-09-04T06:27:28.500240Z", "shell.execute_reply": "2026-09-04T06:27:28.499795Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(figsize=(8, 4.5))\n", "display_poly = poly_table.sort_values(\"cross_scaler_ARI\")\n", "bars = ax.barh(\n", " display_poly.representation.str.replace(\"_\", \" \"),\n", " display_poly.cross_scaler_ARI,\n", " color=[\"#2563eb\" if value == \"BINARY_PRESENCE\" else \"#94a3b8\" for value in display_poly.representation],\n", ")\n", "ax.axvline(.50, color=\"#dc2626\", linestyle=\"--\", linewidth=1.2, label=\"gate final = 0.50\")\n", "ax.set(\n", " title=\"Mengganti representasi saja belum menyelesaikan sensitivitas k=6\",\n", " xlabel=\"ARI antara StandardScaler dan RobustScaler\",\n", " ylabel=\"\",\n", " xlim=(0, .55),\n", ")\n", "ax.bar_label(bars, fmt=\"%.3f\", padding=3, fontsize=9)\n", "ax.legend(loc=\"lower right\")\n", "ax.spines[[\"top\", \"right\"]].set_visible(False)\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "id": "cb35066f", "metadata": {}, "source": [ "Keputusan semantik adalah mengganti `log1p_polyclinic_per_10000_population` dengan `has_polyclinic`. Presence tetap hanya proxy ketersediaan—bukan kapasitas, kualitas layanan, atau akses fisik. Grafik juga menunjukkan bahwa perubahan representasi saja belum cukup; `k` dan keseluruhan pipeline harus dievaluasi ulang." ] }, { "cell_type": "markdown", "id": "b4643f50", "metadata": {}, "source": [ "## 4 — Ruang eksperimen dan hard gate" ] }, { "cell_type": "code", "execution_count": 9, "id": "fb9c4124", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.502691Z", "iopub.status.busy": "2026-09-04T06:27:28.502476Z", "iopub.status.idle": "2026-09-04T06:27:28.516147Z", "shell.execute_reply": "2026-09-04T06:27:28.515612Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
feature_setscaleralgorithmconfigurationsk_mink_max
0A_CURRENT_8ROBUSTAGGLOMERATIVE_WARD3038
1A_CURRENT_8ROBUSTKMEANS3038
2A_CURRENT_8STANDARDAGGLOMERATIVE_WARD3038
3A_CURRENT_8STANDARDKMEANS3038
4B_NO_POLYCLINIC_7ROBUSTAGGLOMERATIVE_WARD2438
5B_NO_POLYCLINIC_7ROBUSTKMEANS2438
6B_NO_POLYCLINIC_7STANDARDAGGLOMERATIVE_WARD2438
7B_NO_POLYCLINIC_7STANDARDKMEANS2438
8C_BINARY_POLYCLINIC_8ROBUSTAGGLOMERATIVE_WARD3038
9C_BINARY_POLYCLINIC_8ROBUSTKMEANS3038
10C_BINARY_POLYCLINIC_8STANDARDAGGLOMERATIVE_WARD3038
11C_BINARY_POLYCLINIC_8STANDARDKMEANS3038
12D_REDUCED_NO_POLY_NO_SEX_6ROBUSTAGGLOMERATIVE_WARD1838
13D_REDUCED_NO_POLY_NO_SEX_6ROBUSTKMEANS1838
14D_REDUCED_NO_POLY_NO_SEX_6STANDARDAGGLOMERATIVE_WARD1838
15D_REDUCED_NO_POLY_NO_SEX_6STANDARDKMEANS1838
16E_BINARY_POLY_NO_SEX_7ROBUSTAGGLOMERATIVE_WARD2438
17E_BINARY_POLY_NO_SEX_7ROBUSTKMEANS2438
18E_BINARY_POLY_NO_SEX_7STANDARDAGGLOMERATIVE_WARD2438
19E_BINARY_POLY_NO_SEX_7STANDARDKMEANS2438
\n", "
" ], "text/plain": [ " feature_set scaler algorithm configurations \\\n", "0 A_CURRENT_8 ROBUST AGGLOMERATIVE_WARD 30 \n", "1 A_CURRENT_8 ROBUST KMEANS 30 \n", "2 A_CURRENT_8 STANDARD AGGLOMERATIVE_WARD 30 \n", "3 A_CURRENT_8 STANDARD KMEANS 30 \n", "4 B_NO_POLYCLINIC_7 ROBUST AGGLOMERATIVE_WARD 24 \n", "5 B_NO_POLYCLINIC_7 ROBUST KMEANS 24 \n", "6 B_NO_POLYCLINIC_7 STANDARD AGGLOMERATIVE_WARD 24 \n", "7 B_NO_POLYCLINIC_7 STANDARD KMEANS 24 \n", "8 C_BINARY_POLYCLINIC_8 ROBUST AGGLOMERATIVE_WARD 30 \n", "9 C_BINARY_POLYCLINIC_8 ROBUST KMEANS 30 \n", "10 C_BINARY_POLYCLINIC_8 STANDARD AGGLOMERATIVE_WARD 30 \n", "11 C_BINARY_POLYCLINIC_8 STANDARD KMEANS 30 \n", "12 D_REDUCED_NO_POLY_NO_SEX_6 ROBUST AGGLOMERATIVE_WARD 18 \n", "13 D_REDUCED_NO_POLY_NO_SEX_6 ROBUST KMEANS 18 \n", "14 D_REDUCED_NO_POLY_NO_SEX_6 STANDARD AGGLOMERATIVE_WARD 18 \n", "15 D_REDUCED_NO_POLY_NO_SEX_6 STANDARD KMEANS 18 \n", "16 E_BINARY_POLY_NO_SEX_7 ROBUST AGGLOMERATIVE_WARD 24 \n", "17 E_BINARY_POLY_NO_SEX_7 ROBUST KMEANS 24 \n", "18 E_BINARY_POLY_NO_SEX_7 STANDARD AGGLOMERATIVE_WARD 24 \n", "19 E_BINARY_POLY_NO_SEX_7 STANDARD KMEANS 24 \n", "\n", " k_min k_max \n", "0 3 8 \n", "1 3 8 \n", "2 3 8 \n", "3 3 8 \n", "4 3 8 \n", "5 3 8 \n", "6 3 8 \n", "7 3 8 \n", "8 3 8 \n", "9 3 8 \n", "10 3 8 \n", "11 3 8 \n", "12 3 8 \n", "13 3 8 \n", "14 3 8 \n", "15 3 8 \n", "16 3 8 \n", "17 3 8 \n", "18 3 8 \n", "19 3 8 " ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "experiment_design = (\n", " experiments.groupby([\"feature_set\", \"scaler\", \"algorithm\"])\n", " .agg(configurations=(\"model_id\", \"count\"), k_min=(\"k\", \"min\"), k_max=(\"k\", \"max\"))\n", " .reset_index()\n", ")\n", "experiment_design" ] }, { "cell_type": "code", "execution_count": 10, "id": "cd0b0b97", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.518772Z", "iopub.status.busy": "2026-09-04T06:27:28.518516Z", "iopub.status.idle": "2026-09-04T06:27:28.534688Z", "shell.execute_reply": "2026-09-04T06:27:28.534222Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
model_idfeature_setksilhouettebootstrap_ARIscaler_robustnessfeature_ablation_robustnesspipeline_consensusparsimony_scorerobustness_scorepasses_gateselected_final
role
Highest raw scoreT05B_D_REDUCED_NO_POLY_NO_SEX_6_KM_STANDARD_PCA5_K6D_REDUCED_NO_POLY_NO_SEX_660.3240.9900.2600.7280.5510.7250.720FalseFalse
V2 selectedT05B_C_BINARY_POLYCLINIC_8_KM_STANDARD_PCA7_K3C_BINARY_POLYCLINIC_830.1760.8280.5490.6220.6730.7000.677TrueTrue
V1 baselineT05B_A_CURRENT_8_KM_STANDARD_PCA6_K6A_CURRENT_860.2220.9380.2700.6770.5390.4250.663FalseFalse
\n", "
" ], "text/plain": [ " model_id \\\n", "role \n", "Highest raw score T05B_D_REDUCED_NO_POLY_NO_SEX_6_KM_STANDARD_PCA5_K6 \n", "V2 selected T05B_C_BINARY_POLYCLINIC_8_KM_STANDARD_PCA7_K3 \n", "V1 baseline T05B_A_CURRENT_8_KM_STANDARD_PCA6_K6 \n", "\n", " feature_set k silhouette bootstrap_ARI \\\n", "role \n", "Highest raw score D_REDUCED_NO_POLY_NO_SEX_6 6 0.324 0.990 \n", "V2 selected C_BINARY_POLYCLINIC_8 3 0.176 0.828 \n", "V1 baseline A_CURRENT_8 6 0.222 0.938 \n", "\n", " scaler_robustness feature_ablation_robustness \\\n", "role \n", "Highest raw score 0.260 0.728 \n", "V2 selected 0.549 0.622 \n", "V1 baseline 0.270 0.677 \n", "\n", " pipeline_consensus parsimony_score robustness_score \\\n", "role \n", "Highest raw score 0.551 0.725 0.720 \n", "V2 selected 0.673 0.700 0.677 \n", "V1 baseline 0.539 0.425 0.663 \n", "\n", " passes_gate selected_final \n", "role \n", "Highest raw score False False \n", "V2 selected True True \n", "V1 baseline False False " ] }, "execution_count": 10, "metadata": {}, "output_type": "execute_result" } ], "source": [ "finalists[\"passes_gate\"] = (\n", " finalists.algorithm.eq(\"KMEANS\")\n", " & finalists.min_cluster_size.ge(15)\n", " & finalists.interpretability_rating.isin([\"HIGH\", \"MEDIUM\"])\n", " & finalists.feature_set.ne(\"A_CURRENT_8\")\n", " & finalists.scaler_robustness.ge(.50)\n", " & finalists.pipeline_consensus.ge(.60)\n", ")\n", "\n", "top_score = finalists.nlargest(1, \"robustness_score\").iloc[0]\n", "decision_rows = finalists.loc[\n", " finalists.model_id.isin([v1_candidate_id, v2_candidate_id, top_score.model_id]),\n", " [\n", " \"model_id\", \"feature_set\", \"k\", \"silhouette\", \"bootstrap_ARI\",\n", " \"scaler_robustness\", \"feature_ablation_robustness\", \"pipeline_consensus\",\n", " \"parsimony_score\", \"robustness_score\", \"passes_gate\", \"selected_final\",\n", " ],\n", "].copy()\n", "decision_rows[\"role\"] = decision_rows.model_id.map({\n", " v1_candidate_id: \"V1 baseline\",\n", " v2_candidate_id: \"V2 selected\",\n", " top_score.model_id: \"Highest raw score\",\n", "})\n", "decision_rows.set_index(\"role\").round(3)" ] }, { "cell_type": "code", "execution_count": 11, "id": "48d19d2a", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.537335Z", "iopub.status.busy": "2026-09-04T06:27:28.537115Z", "iopub.status.idle": "2026-09-04T06:27:28.864709Z", "shell.execute_reply": "2026-09-04T06:27:28.864201Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, ax = plt.subplots(figsize=(9, 6))\n", "failed = finalists.loc[~finalists.passes_gate]\n", "passed = finalists.loc[finalists.passes_gate]\n", "ax.scatter(failed.silhouette, failed.scaler_robustness, color=\"#cbd5e1\", s=55, label=\"Tidak lolos gate\")\n", "ax.scatter(passed.silhouette, passed.scaler_robustness, color=\"#2563eb\", s=70, label=\"Lolos gate\")\n", "ax.scatter(v2_row.silhouette, v2_row.scaler_robustness, color=\"#dc2626\", marker=\"*\", s=230, label=\"V2 terpilih\")\n", "ax.axhline(.50, color=\"#475569\", linestyle=\"--\", linewidth=1, label=\"cross-scaler gate\")\n", "ax.annotate(\"V1\", (v1_row.silhouette, v1_row.scaler_robustness), xytext=(8, -13), textcoords=\"offset points\")\n", "ax.annotate(\"V2\", (v2_row.silhouette, v2_row.scaler_robustness), xytext=(8, 7), textcoords=\"offset points\")\n", "ax.annotate(\"raw score tertinggi (gagal gate)\", (top_score.silhouette, top_score.scaler_robustness), xytext=(8, 7), textcoords=\"offset points\")\n", "ax.set(\n", " title=\"Silhouette yang lebih tinggi tidak otomatis lebih defensible\",\n", " xlabel=\"Silhouette (lebih tinggi = separation internal lebih baik)\",\n", " ylabel=\"Cross-scaler ARI (lebih tinggi = lebih robust)\",\n", ")\n", "ax.legend(loc=\"best\")\n", "ax.spines[[\"top\", \"right\"]].set_visible(False)\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "id": "62416830", "metadata": {}, "source": [ "Kandidat raw-score tertinggi menggunakan enam fitur dan `k=6`, tetapi cross-scaler ARI serta pipeline consensus-nya berada di bawah gate. V2 adalah kandidat ber-score tertinggi setelah syarat berikut diterapkan:\n", "\n", "- K-Means dan minimum cluster size ≥15;\n", "- interpretability `HIGH` atau `MEDIUM`;\n", "- tidak mempertahankan representasi poliklinik V1;\n", "- cross-scaler ARI ≥0,50;\n", "- pipeline consensus ≥0,60.\n", "\n", "Dengan kata lain, pemilihan final adalah **constrained model selection**, bukan leaderboard satu metrik." ] }, { "cell_type": "markdown", "id": "545e0127", "metadata": {}, "source": [ "## 5 — Trade-off V1 versus V2" ] }, { "cell_type": "code", "execution_count": 12, "id": "21dbdc78", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.866940Z", "iopub.status.busy": "2026-09-04T06:27:28.866761Z", "iopub.status.idle": "2026-09-04T06:27:28.877774Z", "shell.execute_reply": "2026-09-04T06:27:28.877314Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
V1V2
metric
Jumlah cluster6.0003.000
Jumlah fitur8.0008.000
Komponen PCA6.0007.000
Silhouette ↑0.2220.176
Davies–Bouldin ↓1.3741.857
Calinski–Harabasz ↑86.12483.430
Bootstrap stability ARI ↑0.9390.839
Cross-scaler ARI ↑0.2700.549
Feature-ablation robustness ↑0.6770.622
Pipeline consensus ↑0.5390.673
Cluster terkecil37.00060.000
Cluster terbesar100.000163.000
\n", "
" ], "text/plain": [ " V1 V2\n", "metric \n", "Jumlah cluster 6.000 3.000\n", "Jumlah fitur 8.000 8.000\n", "Komponen PCA 6.000 7.000\n", "Silhouette ↑ 0.222 0.176\n", "Davies–Bouldin ↓ 1.374 1.857\n", "Calinski–Harabasz ↑ 86.124 83.430\n", "Bootstrap stability ARI ↑ 0.939 0.839\n", "Cross-scaler ARI ↑ 0.270 0.549\n", "Feature-ablation robustness ↑ 0.677 0.622\n", "Pipeline consensus ↑ 0.539 0.673\n", "Cluster terkecil 37.000 60.000\n", "Cluster terbesar 100.000 163.000" ] }, "execution_count": 12, "metadata": {}, "output_type": "execute_result" } ], "source": [ "comparison = pd.DataFrame({\n", " \"metric\": [\n", " \"Jumlah cluster\", \"Jumlah fitur\", \"Komponen PCA\", \"Silhouette ↑\",\n", " \"Davies–Bouldin ↓\", \"Calinski–Harabasz ↑\", \"Bootstrap stability ARI ↑\",\n", " \"Cross-scaler ARI ↑\", \"Feature-ablation robustness ↑\", \"Pipeline consensus ↑\",\n", " \"Cluster terkecil\", \"Cluster terbesar\",\n", " ],\n", " \"V1\": [\n", " v1_config[\"k\"], len(v1_config[\"features\"]), v1_config[\"number_of_pca_components\"],\n", " v1_config[\"metrics\"][\"silhouette\"], v1_config[\"metrics\"][\"davies_bouldin\"],\n", " v1_config[\"metrics\"][\"calinski_harabasz\"], v1_config[\"metrics\"][\"bootstrap_mean_ari\"],\n", " v1_row.scaler_robustness, v1_row.feature_ablation_robustness, v1_row.pipeline_consensus,\n", " v1.cluster_id.value_counts().min(), v1.cluster_id.value_counts().max(),\n", " ],\n", " \"V2\": [\n", " v2_config[\"k\"], len(v2_config[\"features\"]), v2_config[\"pca_components\"],\n", " v2_config[\"metrics\"][\"silhouette\"], v2_config[\"metrics\"][\"davies_bouldin\"],\n", " v2_config[\"metrics\"][\"calinski_harabasz\"], v2_config[\"metrics\"][\"bootstrap_stability_ari\"],\n", " v2_config[\"metrics\"][\"scaler_robustness\"], v2_config[\"metrics\"][\"feature_ablation_robustness\"],\n", " v2_config[\"metrics\"][\"pipeline_consensus\"], v2.cluster_id.value_counts().min(),\n", " v2.cluster_id.value_counts().max(),\n", " ],\n", "})\n", "comparison.set_index(\"metric\").round(3)" ] }, { "cell_type": "code", "execution_count": 13, "id": "951eff96", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:28.879926Z", "iopub.status.busy": "2026-09-04T06:27:28.879716Z", "iopub.status.idle": "2026-09-04T06:27:29.294739Z", "shell.execute_reply": "2026-09-04T06:27:29.294178Z" } }, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(1, 2, figsize=(12, 4.5), sharex=True)\n", "v1_sizes = v1.cluster_id.value_counts().sort_index()\n", "v2_sizes = v2.cluster_label.value_counts().reindex([\"Typology A\", \"Typology B\", \"Typology C\"])\n", "\n", "bars1 = axes[0].barh([f\"Cluster {i}\" for i in v1_sizes.index], v1_sizes.values, color=\"#94a3b8\")\n", "axes[0].bar_label(bars1, padding=3)\n", "axes[0].set(title=\"V1 — enam cluster\", xlabel=\"Jumlah desa/kelurahan\", xlim=(0, 180))\n", "\n", "bars2 = axes[1].barh(v2_sizes.index, v2_sizes.values, color=[COLORS[\"A\"], COLORS[\"B\"], COLORS[\"C\"]])\n", "axes[1].bar_label(bars2, padding=3)\n", "axes[1].set(title=\"V2 — tiga tipologi\", xlabel=\"Jumlah desa/kelurahan\", xlim=(0, 180))\n", "\n", "for axis in axes:\n", " axis.spines[[\"top\", \"right\"]].set_visible(False)\n", "plt.tight_layout()" ] }, { "cell_type": "markdown", "id": "29bd88b1", "metadata": {}, "source": [ "Trade-off-nya eksplisit: V2 menerima separation internal dan bootstrap stability yang lebih rendah daripada V1, sebagai imbalan atas robustness lintas-scaler yang jauh lebih baik, semantics fitur yang lebih aman, pipeline consensus yang lebih kuat, dan struktur yang lebih mudah dikomunikasikan." ] }, { "cell_type": "markdown", "id": "bce084e6", "metadata": {}, "source": [ "## 6 — Bagaimana assignment berubah?" ] }, { "cell_type": "code", "execution_count": 14, "id": "d008cc4f", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:29.296866Z", "iopub.status.busy": "2026-09-04T06:27:29.296642Z", "iopub.status.idle": "2026-09-04T06:27:29.538594Z", "shell.execute_reply": "2026-09-04T06:27:29.537997Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
cluster_labelTypology ATypology BTypology C
cluster_id
119021
202278
321060
40352
53910
60742
\n", "
" ], "text/plain": [ "cluster_label Typology A Typology B Typology C\n", "cluster_id \n", "1 19 0 21\n", "2 0 22 78\n", "3 2 10 60\n", "4 0 35 2\n", "5 39 1 0\n", "6 0 74 2" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "merged = v1[[\"village_id\", \"cluster_id\"]].merge(\n", " v2[[\"village_id\", \"cluster_label\"]], on=\"village_id\", validate=\"one_to_one\"\n", ")\n", "transition = pd.crosstab(merged.cluster_id, merged.cluster_label)\n", "\n", "fig, ax = plt.subplots(figsize=(8, 5))\n", "sns.heatmap(transition, annot=True, fmt=\"d\", cmap=\"Blues\", linewidths=.5, ax=ax)\n", "ax.set(\n", " title=\"V2 bukan sekadar mengganti nama cluster V1\",\n", " xlabel=\"V2\",\n", " ylabel=\"V1 cluster\",\n", ")\n", "plt.tight_layout()\n", "\n", "transition" ] }, { "cell_type": "code", "execution_count": 15, "id": "46f7247e", "metadata": { "execution": { "iopub.execute_input": "2026-09-04T06:27:29.541032Z", "iopub.status.busy": "2026-09-04T06:27:29.540782Z", "iopub.status.idle": "2026-09-04T06:27:29.550640Z", "shell.execute_reply": "2026-09-04T06:27:29.550189Z" } }, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
nilai
ARI V1–V20.337
NMI V1–V20.467
changed after maximum-overlap matching174.000
\n", "
" ], "text/plain": [ " nilai\n", "ARI V1–V2 0.337\n", "NMI V1–V2 0.467\n", "changed after maximum-overlap matching 174.000" ] }, "execution_count": 15, "metadata": {}, "output_type": "execute_result" } ], "source": [ "pd.Series({\n", " \"ARI V1–V2\": adjusted_rand_score(merged.cluster_id, merged.cluster_label),\n", " \"NMI V1–V2\": normalized_mutual_info_score(merged.cluster_id, merged.cluster_label),\n", " \"changed after maximum-overlap matching\": v2_config[\"comparison_to_v1\"][\"changed_after_cluster_matching\"],\n", "}).round(3).to_frame(\"nilai\")" ] }, { "cell_type": "markdown", "id": "adbdfef6", "metadata": {}, "source": [ "Heatmap menunjukkan bahwa V2 bukan penggabungan satu-ke-satu dari enam cluster lama. Karena jumlah cluster dan representasi fitur berubah, ARI/NMI dipakai untuk membandingkan struktur partisi. Angka “changed after matching” hanya alat diagnostik berbasis maximum overlap; ia tidak berarti 174 desa sebelumnya “salah”." ] }, { "cell_type": "markdown", "id": "bc40d16b", "metadata": {}, "source": [ "## 7 — Kesimpulan keputusan model" ] }, { "cell_type": "markdown", "id": "28edb2ea", "metadata": {}, "source": [ "**Mengapa V2 dipilih:**\n", "\n", "1. V1 cukup stabil dalam pengulangan yang sama, tetapi rentan terhadap preprocessing yang wajar.\n", "2. Fitur poliklinik V1 memiliki IQR nol dan 86,85% zero, sehingga laju per kapita memberi presisi yang kurang tepat secara semantik.\n", "3. Sebanyak 504 konfigurasi diuji; 25 finalis menerima evaluasi stability dan ablation tambahan.\n", "4. Hard gate mencegah kandidat dengan silhouette atau raw score tinggi menang ketika cross-scaler robustness/pipeline consensus lemah.\n", "5. V2 `k=3` adalah kandidat terkuat yang lolos seluruh gate dan menghasilkan tipologi yang lebih parsimonious.\n", "\n", "**Yang tidak boleh disimpulkan:** V2 bukan bukti adanya tepat tiga “jenis desa alami”, bukan ranking pembangunan, dan bukan klasifikasi resmi. Assignment tetap memiliki 54 unit dengan ambiguitas tinggi; ketidakpastian tersebut dipublikasikan, bukan disembunyikan." ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.12.6" }, "pasuruan_lens": { "slug": "model-evolution-v1-to-v2", "type": "portfolio_poc" } }, "nbformat": 4, "nbformat_minor": 5 }