Merge pull request #50564 from ShawnKupfer/WB1796

AB#1016252: Fix graphing code in Confusion matrix and data imbalances

Author: JillGrant615

learn-pr/azure/machine-learning-confusion-matrix/notebooks/8-3-exercise-build-matrix.ipynb

@@ -57,10 +57,19 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "import graphing # custom graphing code. See our GitHub repo for details\n",
+    "import matplotlib.pyplot as plt\n",
     "\n",
-    "# Plot a histogram with counts for each label\n",
-    "graphing.multiple_histogram(dataset, label_x=\"label\", label_group=\"label\", title=\"Label distribution\")"
+    "# Count the number of occurrences of each label\n",
+    "label_counts = dataset['label'].value_counts()\n",
+    "\n",
+    "# Plot the histogram\n",
+    "label_counts.plot(kind='bar')\n",
+    "\n",
+    "plt.title(\"Label distribution\")\n",
+    "plt.xlabel(\"Label\")\n",
+    "plt.ylabel(\"Count\")\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -86,7 +95,16 @@
    "outputs": [],
    "source": [
     "# Plot a histogram with counts for each label\n",
-    "graphing.multiple_histogram(dataset, label_x=\"color\", label_group=\"color\", title=\"Color distribution\")"
+    "color_counts = dataset['color'].value_counts()\n",
+    "\n",
+    "# Plot the histogram\n",
+    "color_counts.plot(kind='bar', color='skyblue', edgecolor='black')\n",
+    "\n",
+    "plt.title(\"Color distribution\")\n",
+    "plt.xlabel(\"Color\")\n",
+    "plt.ylabel(\"Count\")\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -110,7 +128,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "graphing.box_and_whisker(dataset, label_y=\"size\", title='Boxplot of \"size\" feature')"
+    "plt.boxplot(dataset['size'].dropna(), vert=True)\n",
+    "\n",
+    "plt.title('Boxplot of \"size\" feature')\n",
+    "plt.ylabel('Size')\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -131,7 +154,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "graphing.box_and_whisker(dataset, label_y=\"roughness\", title='Boxplot of \"roughness\" feature')"
+    "plt.boxplot(dataset['roughness'].dropna(), vert=True)\n",
+    "\n",
+    "plt.title('Boxplot of \"roughness\" feature')\n",
+    "plt.ylabel('Roughness')\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -151,7 +179,12 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "graphing.box_and_whisker(dataset, label_y=\"motion\", title='Boxplot of \"motion\" feature')"
+    "plt.boxplot(dataset['motion'].dropna(), vert=True)\n",
+    "\n",
+    "plt.title('Boxplot of \"motion\" feature')\n",
+    "plt.ylabel('Motion')\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {
@@ -325,41 +358,40 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "# We use plotly to create plots and charts\n",
-    "import plotly.figure_factory as ff\n",
-    "\n",
-    "# Create the list of unique labels in the test set, to use in our plot\n",
-    "# I.e., ['animal', 'hiker', 'rock', 'tree']\n",
-    "x = y = sorted(list(test[\"label\"].unique()))\n",
-    "\n",
-    "# Plot the matrix above as a heatmap with annotations (values) in its cells\n",
-    "fig = ff.create_annotated_heatmap(cm, x, y)\n",
-    "\n",
-    "# Set titles and ordering\n",
-    "fig.update_layout(  title_text=\"<b>Confusion matrix</b>\", \n",
-    "                    yaxis = dict(categoryorder = \"category descending\"))\n",
-    "\n",
-    "fig.add_annotation(dict(font=dict(color=\"black\",size=14),\n",
-    "                        x=0.5,\n",
-    "                        y=-0.15,\n",
-    "                        showarrow=False,\n",
-    "                        text=\"Predicted label\",\n",
-    "                        xref=\"paper\",\n",
-    "                        yref=\"paper\"))\n",
-    "\n",
-    "fig.add_annotation(dict(font=dict(color=\"black\",size=14),\n",
-    "                        x=-0.15,\n",
-    "                        y=0.5,\n",
-    "                        showarrow=False,\n",
-    "                        text=\"Actual label\",\n",
-    "                        textangle=-90,\n",
-    "                        xref=\"paper\",\n",
-    "                        yref=\"paper\"))\n",
-    "\n",
-    "# We need margins so the titles fit\n",
-    "fig.update_layout(margin=dict(t=80, r=20, l=100, b=50))\n",
-    "fig['data'][0]['showscale'] = True\n",
-    "fig.show()"
+    "import numpy as np\n",
+    "\n",
+    "# Create sorted list of unique labels\n",
+    "labels = sorted(list(test[\"label\"].unique()))\n",
+    "\n",
+    "fig, ax = plt.subplots(figsize=(8, 6))\n",
+    "\n",
+    "# Show the confusion matrix\n",
+    "cax = ax.imshow(cm, interpolation='nearest', cmap='Blues')\n",
+    "\n",
+    "# Add colorbar\n",
+    "fig.colorbar(cax)\n",
+    "\n",
+    "# Annotate each cell with the numeric value\n",
+    "for i in range(len(cm)):\n",
+    "    for j in range(len(cm[i])):\n",
+    "        ax.text(j, i, format(cm[i][j], 'd'),\n",
+    "                ha=\"center\", va=\"center\",\n",
+    "                color=\"black\")\n",
+    "\n",
+    "# Set axis labels and ticks\n",
+    "ax.set_xticks(np.arange(len(labels)))\n",
+    "ax.set_yticks(np.arange(len(labels)))\n",
+    "ax.set_xticklabels(labels)\n",
+    "ax.set_yticklabels(labels)\n",
+    "\n",
+    "# Set title and axis labels\n",
+    "plt.title(\"Confusion matrix\", fontsize=14, fontweight='bold')\n",
+    "plt.xlabel(\"Predicted label\", fontsize=12)\n",
+    "plt.ylabel(\"Actual label\", fontsize=12)\n",
+    "\n",
+    "# Adjust layout to fit labels\n",
+    "plt.tight_layout()\n",
+    "plt.show()"
    ]
   },
   {