From 0e28ed6dd08812fd6ae110a46699bc5e2d23b25a Mon Sep 17 00:00:00 2001
From: nuluh <dam.ar@outlook.com>
Date: Mon, 28 Jul 2025 16:41:54 +0700
Subject: [PATCH] feat(notebooks): add cross-dataset validation for Sensor A
 and Sensor B models

Closes #74
---
 code/notebooks/stft.ipynb | 188 ++++++++++++++++++++++++++++++++++++++
 1 file changed, 188 insertions(+)

diff --git a/code/notebooks/stft.ipynb b/code/notebooks/stft.ipynb
index ceb9176..4510922 100644
--- a/code/notebooks/stft.ipynb
+++ b/code/notebooks/stft.ipynb
@@ -938,6 +938,194 @@
     "print(\"Accuracy on Dataset B:\", accuracy_score(y, y_pred_svm))\n",
     "print(classification_report(y, y_pred_svm))"
    ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "## Cross Dataset Validation"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "---"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Training"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.model_selection import train_test_split\n",
+    "\n",
+    "# sensor A\n",
+    "x_train1, x_test1, y_train1, y_test1 = train_test_split(X1b, y, test_size=0.2, random_state=2)\n",
+    "# sensor B\n",
+    "x_train2, x_test2, y_train2, y_test2 = train_test_split(X2b, y, test_size=0.2, random_state=2)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "results_sensor1 = []\n",
+    "for name, model in models_sensor1.items():\n",
+    "    res = train_and_evaluate_model(model, name, \"Sensor A\", x_train1, y_train1, x_test1, y_test1, \n",
+    "                                   export='D:/thesis/datasetB/models/Sensor A')\n",
+    "    results_sensor1.append(res)\n",
+    "    print(f\"{name} on sensor1: Accuracy = {res['accuracy']:.2f}%\")\n",
+    "\n",
+    "# Display result\n",
+    "results_sensor1"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "results_sensor2 = []\n",
+    "for name, model in models_sensor2.items():\n",
+    "    res = train_and_evaluate_model(model, name, \"sensor2\", x_train2, y_train2, x_test2, y_test2, \n",
+    "                                   export='D:/thesis/datasetB/models/sensor2')\n",
+    "    results_sensor2.append(res)\n",
+    "    print(f\"{name} on sensor2: Accuracy = {res['accuracy']:.2f}%\")\n",
+    "\n",
+    "# Display result\n",
+    "results_sensor2"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Testing"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Sensor A"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# 4. Sensor A Validate on Dataset A\n",
+    "from joblib import load\n",
+    "svm_model = load('D:/thesis/datasetB/models/sensor1/SVM.joblib')\n",
+    "y_pred_svm_1 = svm_model.predict(X1a)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from sklearn.metrics import accuracy_score, classification_report\n",
+    "\n",
+    "# 5. Evaluate\n",
+    "print(\"Accuracy on Dataset B:\", accuracy_score(y, y_pred_svm_1))\n",
+    "print(classification_report(y, y_pred_svm_1))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Confusion Matrix Sensor A"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
+    "\n",
+    "\n",
+    "cm = confusion_matrix(y, y_pred_svm_1) # -> ndarray\n",
+    "\n",
+    "# get the class labels\n",
+    "labels = svm_model.classes_\n",
+    "\n",
+    "# Plot\n",
+    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=labels)\n",
+    "disp.plot(cmap=plt.cm.Blues)  # You can change colormap\n",
+    "plt.title(\"Confusion Matrix of Sensor A Test on Dataset B\")\n",
+    "plt.show()"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Sensor B"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "svm_model = load('D:/thesis/datasetB/models/sensor2/SVM.joblib')\n",
+    "# svm_model = load('D:/thesis/models/sensor2/SVM with StandardScaler and PCA.joblib')\n",
+    "y_pred_svm_2 = svm_model.predict(X2a)\n",
+    "\n",
+    "# 5. Evaluate\n",
+    "print(\"Accuracy on Dataset B:\", accuracy_score(y, y_pred_svm_2))\n",
+    "print(classification_report(y, y_pred_svm_2))"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "#### Confusion Matrix Sensor B"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import matplotlib.pyplot as plt\n",
+    "from sklearn.metrics import confusion_matrix, ConfusionMatrixDisplay\n",
+    "\n",
+    "\n",
+    "cm = confusion_matrix(y, y_pred_svm_2) # -> ndarray\n",
+    "\n",
+    "# get the class labels\n",
+    "labels = svm_model.classes_\n",
+    "\n",
+    "# Plot\n",
+    "disp = ConfusionMatrixDisplay(confusion_matrix=cm, display_labels=labels)\n",
+    "disp.plot(cmap=plt.cm.Blues)  # You can change colormap\n",
+    "plt.title(\"Confusion Matrix of Sensor B Test on Dataset B\")\n",
+    "plt.show()"
+   ]
   }
  ],
  "metadata": {