accuracy metric fixed

Author: jordanlei

analysis/Demo Analysis.ipynb

@@ -0,0 +1,68 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 18,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import numpy as np"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 26,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "a = np.array([1, 2, 2]) \n",
+    "b = np.array([2, 2, 2])"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "0.5"
+      ]
+     },
+     "execution_count": 29,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "def equal_weight_acc(truth, predictions): \n",
+    "    indices = [truth == x for x in set(truth)]\n",
+    "    return np.mean([(truth[i] == predictions[i]).mean() for i in indices])\n",
+    "\n",
+    "equal_weight_acc(a, b)"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python (jordan_env)",
+   "language": "python",
+   "name": "jordan_env"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.7.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

analysis/Untitled.ipynb

@@ -92,14 +92,14 @@ def plot_boxplots(df, save = False):
     plt.figure(figsize = (10, 5))
     sns.set_context("poster")
     sns.barplot(data = df, x = "penalty", y = "train_acc", capsize=.2)
-    plt.xlabel("Gating Loss Penalty")
+    plt.xlabel("Loss Parameter Φ")
     plt.ylabel("Training Accuracy")
     plt.ylim([0.8, 1.0])
     plt.savefig("graphs/train_boxplot.svg")
 
     plt.figure(figsize = (10, 5))
     sns.barplot(data = df, x = "penalty", y = "final_acc", capsize=.2)
-    plt.xlabel("Gating Loss Penalty")
+    plt.xlabel("Loss Parameter Φ")
     plt.ylim([0.8, 1.0])
     plt.ylabel("Validation Accuracy")
     plt.savefig("graphs/val_boxplot.svg")