Added line for actual weight trend line

This is in response to pull request #257. That hard coded the
line, which meant it had no label in the legend. This adds the
trend line with a label.

01-g-h-filter.ipynb

@@ -38,7 +38,7 @@
        "        <style>\n",
        "        .output_wrapper, .output {\n",
        "            height:auto !important;\n",
-       "            max-height:10000px; \n",
+       "            max-height:100000px; \n",
        "        }\n",
        "        .output_scroll {\n",
        "            box-shadow:none !important;\n",
@@ -245,7 +245,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Average of measurements is 165.0204\n"
+      "Average of measurements is 165.0040\n"
      ]
     }
    ],
@@ -276,7 +276,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Average of measurements is 164.9414\n"
+      "Average of measurements is 165.0454\n"
      ]
     }
    ],
@@ -608,30 +608,29 @@
     "time_step = 1.0  # day\n",
     "scale_factor = 4.0/10\n",
     "\n",
-    "def predict_using_gain_guess(estimate, gain_rate, do_print=False):     \n",
-    "    # store the filtered results\n",
-    "    estimates, predictions = [estimate], []\n",
+    "def predict_using_gain_guess(estimated_weight, gain_rate, do_print=False):     \n",
+    "    # storage for the filtered results\n",
+    "    estimates, predictions = [estimated_weight], []\n",
     "\n",
     "    # most filter literature uses 'z' for measurements\n",
     "    for z in weights: \n",
     "        # predict new position\n",
-    "        prediction = estimate + gain_rate * time_step\n",
+    "        predicted_weight = estimated_weight + gain_rate * time_step\n",
     "\n",
     "        # update filter \n",
-    "        estimate = prediction + scale_factor * (z - prediction)\n",
+    "        estimated_weight = predicted_weight + scale_factor * (z - predicted_weight)\n",
     "\n",
-    "        # save\n",
-    "        estimates.append(estimate)\n",
-    "        predictions.append(prediction)\n",
+    "        # save and log\n",
+    "        estimates.append(estimated_weight)\n",
+    "        predictions.append(predicted_weight)\n",
     "        if do_print:\n",
-    "            gh.print_results(estimates, prediction, estimate)\n",
+    "            gh.print_results(estimates, predicted_weight, estimated_weight)\n",
     "\n",
     "    return estimates, predictions\n",
     "\n",
     "initial_estimate = 160.\n",
-    "\n",
     "estimates, predictions = predict_using_gain_guess(\n",
-    "    estimate=initial_estimate, gain_rate=1, do_print=True)     "
+    "    estimated_weight=initial_estimate, gain_rate=1, do_print=True)     "
    ]
   },
   {
@@ -643,7 +642,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7MAAAGACAYAAAB/bzVFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xd8VFX6+PHPpBMghISENEggYIDQpImAFaWIdAbF3RUsfHf94aq7X9e1g7q6u+oWV92vIrhiZQkiIN2uKFVqEmpISCe9tyn398clNzPJhISQMpP7vF8vXgy3nDl3npnDPHPOPcegKIqCEEIIIYQQQgjhQtw6ugJCCCGEEEIIIcTlkmRWCCGEEEIIIYTLkWRWCCGEEEIIIYTLkWRWCCGEEEIIIYTLkWRWCCGEEEIIIYTLkWRWCCGEEEIIIYTL8ejoCnQ0q9WK1Wq122YwGDAYDB1UIyGEEEIIIYTQF0VRqL9qrJubG25ujfe/SjJrtVJeXt7R1RBCCCGEEEIIYaNr166XTGZlmLEQQgghhBBCCJcjyawQQgghhBBCCJcjyawQQgghhBBCCJej+3tmHU301NTYbCGEEEIIIYS4HGazme+//55Nmzaxfft2ioqK6NevH7NmzWLWrFkMHTqUnJwcVqxYwebNmzGZTNq5Y8aM4dlnn2X8+PEdeAVty9FcRk1NymtQ6k8ZpTNWq5XS0lK7bd27d5dkVgghhBBCCHFFampq+Oqrr4iLi2Pjxo0UFhYycOBAjEYjRqORESNGaAlbZmYmEydOJCUlxWFZnp6ebNy4kdtuu60dr6D9tCQvk2RWklkhhBBCCCFEK6mpqeHLL7/UEtiioiKuuuoqLYEdPny4wx7H+fPns2HDBgB69uzJ4sWLCQgIYO3atSQmJmrbU1NT6datW7teU3uQZLYF9J7Mnj59GrPZjIeHB1dddVVHV0e0A4m5vki89UXirT8Sc32ReDuvmpoavvjiC+Li4ti0aRNFRUXExMRoCeywYcMuOWQ2LS2NqKgorFYrQUFBHDlyhLKyMsxmMwB/+MMf2LZtGwArV65k6dKl7XJd7akleZnu75nVu8rKSkwmE56enh1dFdFOJOb6IvHWF4m3/kjM9UXi7Vyqq6vtEtji4mJiYmJ48MEHWbhwIUOHDm3yns9a3333HVarFYD/+Z//ISwsjKNHj2rxfvrpp7Vk9uuvv+6UyWxLSDKrcwaDQfsj9EFiri8Sb32ReOuPxFxfJN4dr7q6ml27dhEXF8fmzZspLi5m0KBBPPTQQxiNxstKYG1VVVVpj0NCQgD7eIeGhmr7Kysrr/xCOglJZnVu+PDhHV0F0c4k5voi8dYXibf+SMz1ReLdMWoT2HXr1rF582ZKSkoYPHgwDz/8MEajkdjY2Cv+gWHgwIHa47i4OJYtW2YX71deeUV7LEPM68g9szq/Z1YIIYQQQghhr6qqyq4HtqSkhCFDhmj3wMbGxrbq81mtVmJiYjh79iwA9957L48//jgBAQF8+OGHPP7441rv7YkTJxg0aFCrPr8zkAmgWkCSWSGEEEIIIURVVRU7d+7UEtjS0lJiY2O1BHbIkCFt+vwbNmxg/vz5lzzm3nvvZfXq1W1aj44iyWwLSDIrhBBCCCGEPlVVVbFjxw7i4uL4/PPP2z2Bre/dd9/l//2//0d1jQnviFjcu/XEUlZIdXoCi+/+FStXrsTLy6td69ReZDZjcdlyc3OxWCy4u7sTFBTU0dUR7UBiri8Sb32ReOuPxFxfJN6to7Ky0i6BLSsrY+jQoTz66KMYjUYGDx7cYXW799578b1qAi9sPUG5Upe0BnZx4875IzttIttSHd79WFpaymOPPcaUKVMICgrCYDCwYsWKBsfZzuZV/4/tmPHTp0/z6KOPMnr0aPz9/QkICGDixImsX7++Ha/KdWRmZpKenk5mZmZHV0W0E4m5vki89UXirT8Sc32ReLdcZWUlGzZsYNGiRQQHBzNv3jzi4+N57LHHOHHiBMePH+fZZ5/t0EQWYEd8Fk9sSbJLZAEKKq088OEhdsRndVDNnFOH98zm5+ezcuVKRowYwZw5c1i1apXD4/bs2dNg2759+3jkkUeYO3eutm3Xrl1s3bqVX/3qV4wdOxaz2cx///tfjEYjzz33HM8++2ybXYsQQgghhBDCOVRWVrJ9+3bWrVvHli1bKC8vZ/jw4fzxj3/EaDQSExPT0VW0Y7EqPPd5Io7uAVUAA/Dc54ncOiQEdzdZngmc4J7Z2qc3GAzk5eURFBTE8uXLHfbO1nfPPfewZs0aTp8+zYABAwDIy8sjMDCwwfTYt99+O9988w0FBQV4e3tr2/V+z2xRURFWqxU3Nzf8/f07ujqiHUjM9UXirS8Sb/2RmOuLxLtpFRUVbN++nbi4OC2BHTFihHYPrDMva7MnKZ9F7+xt8rhPlo7n2ujAdqhR+3LJe2ZbuiZTaWkpcXFx3HDDDVoiC9CrVy+Hx48bN46tW7dSUFBgt+iw3klDqD8Sc32ReOuLxFt/JOb6IvF2rKKigm3btmkJbEVFBSNHjuTJJ59kwYIFTp3AApgsVr4+mcNrX54mtCSXgIriRo/N9/Unp7SqHWvn3Do8mW2ptWvXUl5ezv3339+s47/55huCgoIIDg5u45oJIYQQQggh2lJ5ebmWwG7dulVLYJ966imMRiMDBw7s6Co2Ka2ggv8eSGPdwTRySqvxMpv4cc3vCKooavScnK49OfdgfDvW0rm5bDK7evVq/P39m1yLCWDVqlV8++23vPbaa7i7uzd5fEJCApGRkfj5+WnbqqurOXnyJAA9e/akb9++duecOXOGiooKAEaMGGG3Ly8vj4yMDAD69u1Lz549tX0Wi4X4ePUN2b17d/r37293bnJyMiUlJQDExsbi4VEXsqKiIs6fPw9AWFhYg1ntjh07hqIodOnSpcEvUmlpaRQUFAAQExODj4+Ptq+srIykpCQAgoODG/RkJyYmYjKZ8PT0bDBdeVZWFjk5OQBER0fTrVs3bV9VVRWnTp0CICAggD59+tide/r0aSorKzEYDAwfPtxuX25urjbZQWRkpN0vk2azmYSEBAD8/Pzo16+f3bnnzp3ThiwMHTrU7j1QWFhIamoqAOHh4Q169o8ePQqAr69vg0YxNTWVwsJCAAYNGmQ3fL2kpITk5GQAevfuTUhIiN25CQkJmM1mvL29Gyx6nZmZSW5uLgADBgyga9eu2r6KigrOnDkDQGBgIBEREXbnnjp1iqqqKtzd3Rk6dKjdvpycHLKy1EkDoqKi6NGjh7avpqaGEydOANCjRw+ioqLszk1KSqKsrAyAYcOG2Q33yM/PJz09HYCIiAgCA+uGvVitVo4fPw5At27diI6Otis3JSWF4mL118fBgwfbzdBXXFxMSkoKAKGhoQ1+iIqPj8diseDj49Pgnpf09HTy8/MBGDhwIL6+vtq+8vJybUHyoKAgwsLC7M49efIk1dXVeHh4NFgQPTs7mwsXLgDQr18/aSOQNkLaCJW0ESppI1TSRtSRNkLVWm1ESEgIW7duJS4ujm3btlFRUcGgQYO47777+M1vfmP3fnLWNiI9M5v9GZXszrSyJ6WI2hs+A7t6cX1EVy74BxFYUYybgztnrRjI7dGLsTH274fO1kZcDpdMZhMSEti3bx/Lli2ze2Ec2b59O8uWLWPBggX89re/bVb5ZrOZ+rcSK4qCyWTS9js6p3Z/fVarVdtntVob7G9uufXrZFuuxWJxWK6iKHh6ejbYZ7FYtHNNJhMeHh7am7c55TZ2rbbl1r9W29fQUbm11+po6Hlzy3WW2DR1rSaTCbPZ7PAeANtrbUm5JpPJ4b76MTebzXYNVktj3tzX0NH5tnWqr6lrrampwWq1OvyBqjVeQ0fTCTS3XGd5H9qW6+i5m9tGtKTtkTaiY9oI2+e8kjbC0ftf2oi6+jpTG2F7Xkd9j5A2ov3aiNrvqDU1NVgslg77HtGebURxcTG7d+/mu+++44cffqCyspJRo0bxzDPPMHbsWC2ptr3tsDnX2t5tRHJeOau+T2NLQh7F1ep1Ggxw3cBe3DWuLxP79eD0yUTOLf01Q5971MErBW4oJC39NbHu9nF39lzjcsu9HC6ZzK5evRqgySHGO3fuZN68edx666189NFHzb4/18PDo8GxBoNBC5RtAmB7jqNAAri5uWn7HDU6zS23fp1sy3X0QfT09ERRFIfluru74+npiclk4tSpU3h6emq/BDenXNu/HZXr6FptX0NH5dZeq6M4NbdcZ4lNU9dae51NvYYtKbd2/bnGyjWZTKSkpNjF3PZaGyv3Sl/Dpq61vqau1cvLC4vF0iavodVqveTnpqlyneV9aFvWhQsXGvx63Jw2orFypY1QOVsbce7cObvjHJXbVBvhqNzacy9VrrQRHdNG1PaGNafctvge0dJybf9urFxpIxpea0JCgpYENPa5a+vvEY7Ktb3W1mgjysrK2Lp1K+vWrWPbtm1UVVUxdOhQli9fzoIFC7RRHLajN+pzhjai2mxhR3w2a/ensedcvnZMzy5uLLg6nLsnDaBPgNrrW11djaenJ5FzJpO3Npaep0/grtQlphaDG/kDYoicM7lZr2H9OoHrtBGXo8NnM7bVnNmMa2pqCA8PJzIykoMHDzZa1s6dO5kzZw433HADmzZtshu6YUvvsxkfPXoUk8nUILERnZfEXF8k3voi8dYfibm+dOZ4l5WVsWXLFm0IcVVVFWPGjMFoNLJgwYIGw2Od2dmcUj7Zn8aGQ+kUVqg/PrgZ4MaYYO4c24ebBwXjYdu7qiiwdy+89x4MGwYDB8K0aQ0L3rEDpk5tn4voAC45m/Hl2rx5M3l5eTz//PONHrNr1y7mzJnDpEmT2LhxY6OJrFDvDak/3FR0bhJzfZF464vEW38k5vrS2eJdWlqqJbDbt2+nqqqKsWPH8vzzz7NgwYIG9y87syqLine truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -655,7 +654,7 @@
    "source": [
     "# plot results\n",
     "book_plots.set_figsize(10)\n",
-    "gh.plot_gh_results(weights, estimates, predictions)"
+    "gh.plot_gh_results(weights, estimates, predictions, [160, 173])"
    ]
   },
   {
@@ -680,7 +679,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA7MAAAGACAYAAAB/bzVFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3Xl8VNXZwPHfZCEhG2QlG0kQAiEJBNlE3HABQUQCyaBWWqyVVl9sa6svUlxAqFZrrVq1tQhaW7GUAGWVxQUXXABZAgkQICRkJ/ueSWa57x95c8kkAwkhySS5z/fz4UO4596Tc+eZe5hn7rnn6BRFURBCCCGEEEIIIXoRB3s3QAghhBBCCCGEuFKSzAohhBBCCCGE6HUkmRVCCCGEEEII0etIMiuEEEIIIYQQoteRZFYIIYQQQgghRK8jyawQQgghhBBCiF7Hyd4NsDeLxYLFYrHaptPp0Ol0dmqREEIIIYQQQmiLoii0XDXWwcEBB4dL33+VZNZioaamxt7NEEIIIYQQQgjRjLu7+2WTWRlmLIQQQgghhBCi15FkVgghhBBCCCFEryPJrBBCCCGEEEKIXkfzz8zamuiprbHZQgghhBBCCHElTCYTX331FVu2bGHnzp2Ul5czZMgQ7rnnHu655x5iY2MpLCxk+fLlbN26FaPRqB47fvx4nnvuOSZNmmTHM+hatuYyamtSXp3ScsoojbFYLFRVVVlt8/T0lGRWCCGEEEIIcVUaGhr47LPPSEpKYvPmzZSVlREZGYler0ev1xMXF6cmbHl5edxwww1kZmbarMvZ2ZnNmzdz1113deMZdJ+O5GWSzEoyK4QQQgghhOgkDQ0NfPrpp2oCW15ezvDhw9UEdvTo0TbvOCYkJLBp0yYAvL29WbBgAT4+Pqxbt44TJ06o27OysvDw8OjWc+oOksx2gNaT2dOnT2MymXBycmL48OH2bo7oBhJzbZF4a4vEW3sk5toi8e65Ghoa+OSTT0hKSmLLli2Ul5czYsQINYEdNWrUZYfMZmdnExERgcViwd/fn6NHj1JdXY3JZALgf//3f/n4448BWLVqFQsXLuyW8+pOHcnLNP/MrNbV1dVhNBpxdna2d1NEN5GYa4vEW1sk3tojMdcWiXfPUl9fb5XAVlRUMGLECB577DHmzZtHbGxsm898Nvnyyy+xWCwA/PznPyc4OJjk5GQ13s8884yazH7++ed9MpntCElmNU6n06l/hDZIzLVF4q0tEm/tkZhri8Tb/urr69mzZw9JSUls3bqViooKoqKi+NWvfoVer7+iBLY5g8Gg/hwYGAhYxzsoKEgtr6uru/oT6SMkmdW40aNH27sJoptJzLVF4q0tEm/tkZhri8TbPpoS2PXr17N161YqKysZOXIkv/71r9Hr9cTExFz1FwyRkZHqz0lJSSxatMgq3q+88or6swwxv0iemdX4M7NCCCGEEEIIawaDweoObGVlJdHR0eozsDExMZ36+ywWCyNGjODs2bMAPPTQQyxZsgQfHx8+/PBDlixZot69PXnyJFFRUZ36+3sCmQCqAySZFUIIIYQQQhgMBnbv3q0msFVVVcTExKgJbHR0dJf+/k2bNpGQkHDZfR566CHWrFnTpe2wF0lmO0CSWSGEEEIIIbTJYDCwa9cukpKS2LZtW7cnsC299957/M///A/19fWtyhYsWMCqVavo169ft7apu0gy2wFaT2aLioowm804Ojri7+9v7+aIbiAx1xaJt7ZIvLVHYq4tEu/OUVdXZ5XAVldXExsbqyawI0eOtGv7Lly4wJo1a/j0009paGhg+PDh/PKXv+Taa6+1a7u6Wq9cmqeqqoqVK1dy9OhRjhw5QnFxMcuWLWP58uVW+13uoeoRI0Zw6tQpoHH9rVWrVrF3717S09NxcHBg5MiR/OY3vyExMbErT6VXysvLU6f8lk5RGyTm2iLx1haJt/ZIzLVF4t1xdXV17Ny5k6SkJLZv3051dTWjRo1i8eLF6PX6HvUM6qBBg1i6dCkzZ85U4x0XF2fvZvVIdk9mS0pKWLVqFXFxccTHx7N69Wqb+3333Xettu3fv5/HH3+cOXPmqNv27NnDjh07+PGPf8yECRMwmUz85z//Qa/X8/zzz/Pcc8912bkIIYQQQggheoamBHb9+vVs376dmpoaRo8ezVNPPYVer2fEiBH2bqK4SnYfZtz063U6HcXFxfj7+9u8M2vLT3/6Uz744ANOnz7NsGHDACguLsbX17fVndy7776bvXv3UlpaiouLi7pd68OMy8vLsVgsODg4MHDgQHs3R3QDibm2SLy1ReKtPRJzbZF4t622ttbqDmxNTQ1xcXHqEOLetKyN1uLdK4cZd3RNpqqqKpKSkrjlllvURBbAz8/P5v4TJ05kx44dlJaWWi06rHVauDCENYm5tki8tUXirT0Sc22ReNtWW1vLxx9/rCawtbW1jBkzhqVLl5KYmNirEtjmJN5ts3sy21Hr1q2jpqaGhx9+uF377927F39/fwICArq4ZUIIIYQQQoiuVFNToyawO3bsUBPYp59+Gr1eT2RkpL2bKLpBr01m16xZw8CBA9tciwlg9erVfPHFF7zxxhs4Ojq2uX9qairh4eF4eXmp2+rr69VJpry9vQkLC7M65syZM9TW1gK0ekC7uLiY3NxcAMLCwvD29lbLzGYzKSkpQONt9Guuucbq2IyMDCorKwGIiYnByeliyMrLyzl//jwAwcHBrSYCOHbsGIqi0L9//1bfSGVnZ1NaWgo0TqDl6uqqllVXV5Oeng5AQEBAqzvZJ06cUB9GbzldeX5+PoWFhQAMHToUDw8PtcxgMJCWlgaAj48PgwcPtjr29OnT1NXVodPpGD16tFVZUVEReXl5AISHh1t9U2UymUhNTQXAy8uLIUOGWB177tw5dchCbGys1XugrKyMrKwsAEJCQlrd2U9OTgbAzc2tVaeYlZVFWVkZAFFRUVbD1ysrK8nIyAAaH+IPDAy0OjY1NRWTyYSLi0urCQfy8vIoKioCYNiwYbi7u6tltbW1nDlzBgBfX19CQ0Otjk1LS8NgMODo6EhsbKxVWWFhIfn5+QBEREQwYMAAtayhoYGTJ08CMGDAACIiIqyOTU9Pp7q6GoBRo0ZZDfcoKSkhJycHgNDQUHx9fdUyi8XC8ePHAfDw8GDo0KFW9WZmZlJRUQHAyJEjraaar6ioIDMzE4CgoKBWX0SlpKRgNptxdXVt9cxLTk4OJSUlAERGRuLm5qaW1dTUqAuS+/v7ExwcbHXsqVOnqK+vx8nJqdWC6AUFBVy4cAGAIUOGSB+B9BHSRzSSPqKR9BGNpI+4SPqIRp3VRwQGBrJjxw6SkpL4+OOPqa2tJSoqip/97Gc88sgjVu8n6SMa9cY+4kr0ymQ2NTWV/fv3s2jRIqsXxpadO3eyaNEiEhMT+eUvf9mu+k0mEy0fJVYUBaPRqJbbOqapvCWLxaKWWSyWVuXtrbdlm5rXazabbdarKArOzs6tysxms3qs0WjEyclJffO2p95LnWvzeluea/PX0Fa9Tedqa+h5e+vtKbFp61yNRiMmk8nmMwDNz7Uj9RqNRptlLWNuMpmsOqyOxry9r6Gt45u3qaW2zrWhoQGLxWLzC6rOeA1tTSfQ3np7yvuweb22fnd7+4iO9D3SR9inj2j+O6+mj7D1/pc+4mJ7e1If0fw4e32OkD6i+/qIps+oDQ0NmM1mu32O6M4+oqKign379vHll1/y9ddfU1dXx9ixY3n22WeZMGGCmlQ3f+ywPefaG/qIpnjrdDq7f47ozj7iSvTKZHbNmjUAbQ4x3r17N3PnzmXq1KmsXbu23c/nOjk5tdpXp9OpgWqeADQ/xlYgARwcHNQyW51Oe+tt2abm9dq6EJ2dnVEUxWa9jo6OODs7YzQaSUtLs5ryuz31Nv/bVr22zrX5a2ir3qZztRWn9tbbU2LT1rk2nWdbr2FH6m1af+5S9RqNRjIzM1tN895WvVf7GrZ1ri21da79+vXDbDZ3yWtosVgue920VW9PeR82r+vChQutvj1uTx9xqXqlj2jU0/qIc+fOWe1nq962+ghb9TYde7l6pY+wTx/RdDesPfV2xeeIjtbb/O9L1St9ROtzTU1NVZOAS113Xf05wla9zc+1M/qI6upqduzYwfr16/n4448xGAzExsaybNkyEhMT1VEczUdvtNQX+oimeDs7O+Pm5tYrc42O1Hsl7D6bcXPtmc24oaGBkJAQwsPD+eGHHy5Z1+7du4mPj+eWW25hy5YtVkM3mtP6bMbJycnqRSLrV2mDxFxbJN7aIvHWHom5tvTleFdXV7N9+3Z1CLHBYGD8+PHo9XoSExNbDY/Vgr4cb1t65WzGV2rr1q0UFxezYsWKS+6zZ88e4uPjufHGG9m8efMlE1nR+GxIy+Gmom+TmGuLxFtbJN7aIzHXlr4W76qqKjWB3blzJwaDgQkTJrBixQoSExNbPb+sNX0t3l2hR9yZ3blzJzU1NVRVVfHQQw+h1+uZN28eAHfddZfVQ9czZszgyy+/JD8/3+rB8yb79u1j2rRpDBo0iPfee4/+/ftblUdHR1s9bK31O7NCCCGEEEJ0l6qqKrZt20ZSUhK7du1SE9imO7BaT2C1rCN5WY9IZiMiItSZslrKyMhQZ0TLzs4mIiKC+fPn88EHH9jcf/ny5Tz//POX/F179+5lypQp6r8lmRVCCCGEEKLrVFZWWiWw9fX1TJw4UU1gW85+LLSp1yaz9iTJrBBCCCGEEJ2rsrKSrVu3kpSUxO7du6mvr+e6665TE9jw8HB7N1H0MJp4ZlYIIYQQQgjR81RUVFjdgW1oaGDSpEm8+OKLJCYmtlo7VYirJcmsxp07d059sFyLs8RpkcRcWyTe2iLx1h6Jubb0xHhXVFRY3YFtaGjg+uuv56WXXiIhIUES2KvQE+Pd00gyq3FVVVXqlN9CGyTm2iLLine truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -691,7 +690,7 @@
    ],
    "source": [
     "e, p = predict_using_gain_guess(initial_estimate, -1.)\n",
-    "gh.plot_gh_results(weights, e, p)"
+    "gh.plot_gh_results(weights, e, p, [160, 173])"
    ]
   },
   {
@@ -725,7 +724,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "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 truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -758,7 +757,7 @@
     "  \n",
     "    estimates.append(weight)\n",
     "\n",
-    "gh.plot_gh_results(weights, estimates, predictions)"
+    "gh.plot_gh_results(weights, estimates, predictions, [160, 173])"
    ]
   },
   {
@@ -1058,7 +1057,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "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 truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -1095,7 +1094,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "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 truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -1149,7 +1148,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA64AAAFrCAYAAADRvNRcAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XlcVNX7wPHPzLDvAoILuAQq7kuaWZlLimaaZu6lZqatLuXXzNI0t7RFszTLpSyz1NzK0swllyx3RcEdkUVQZFdggJm5vz/4cWUYEFQExp736+XLYc69Z86983CYZ+6552gURVEQQgghhBBCCCEqKG15N0AIIYQQQgghhLgVSVyFEEIIIYQQQlRokrgKIYQQQgghhKjQJHEVQgghhBBCCFGhSeIqhBBCCCGEEKJCsynvBtwuk8mEyWQye06j0aDRaMqpRUIIIYQQQgjx36IoCgUXqNFqtWi19+baqFUmrunp6eXdDCGEEEIIIYQQ+Tg7O9+zxFWGCgshhBBCCCGEqNAkcRVCCCGEEEIIUaFJ4iqEEEIIIYQQokKzuntcC5uE6V6OpRZCCCGEEEIIYa6wuYfu5YS590Xiei9nrxJCCCGEEEIIUbx7mbhKtieEEEIIIYQQokKzuiuuQlizc+fOYTAYsLGxoW7duuXdHCFKROJWWCOJW2GNJG6FKJokrkKUoczMTHJycrC1tS3vpghRYhK3whpJ3AprJHErRNFkqLAQZUij0aj/hLAWErfCGkncCmskcStE0TSKoijl3YjbYTKZuH79utlzrq6uMjmTEEIIIYQQQpSRss7LJNsTQgghhBBCCFGhSeIqhBBCCCGEEKJCk8RVCCGEEEIIIUSFJrMKC1GGrl27htFoRKfTUbly5fJujhAlInErrJHErbBGErcVk6IoHDx4kL1792IwGKhfvz7dunW77dmfTSYTR44cIT4+Hi8vL1q2bImNzc107PDhwyxYsIDt27eTmZlJQEAAL774IkOGDMHJyam0D8vqSOIqRBmKjY1Vp7mXP0jCWkjcCmskcSuskcRtxRMSEsLw4cM5cuSI2fNVq1blk08+YdCgQcXWoSgKy5YtY/bs2YSHh6vP+/v7M27cOEaNGsXs2bN57733zPZLSkri0KFDfP755/z555/4+fmVzkFZKUlchRBCCCGEEPelhIQEfvrpJyIiInByciI4OJi2bduWaMmh0NBQHn/8cdLS0izK4uLieO6558jMzGT48OG3rGfChAl8/PHHFs9HR0czduxYNmzYwO7du9XnnZ2d8fT0JDo6GoDTp0/TvXt3Dh8+bHaF9r9GlsMRogylpKRgMpnQarV4eHiUd3OEKBGJW2GNJG6FNZK4LT1Go5HJkyfz6aefkp2dbVbWtGlTfvzxRxo0aHDLOh5//HH27t0LQP369Rk1ahQeHh788MMPbN68GQBHR0diYmLw9PQstI5t27YRHBys/tyxY0cee+wxDh06xJYtWyy2nzJlCm+//TZOTk4cPnyYfv36ERERAcD69et55plnSn4S7tKRI0c4ePAgiqLQrFkz2rRpY5bwl3VeJomrEEIIIYQQ4r4yatQoFixYUGS5t7c3Bw4c4IEHHii0PCwsjEaNGgEQEBBASEgIzs7OQO7Q32HDhvHdd98BMG/ePMaOHVtoPT169OC3334rdLvvv/+eoUOHqj8//vjjZldeAbZs2UK3bt0A6N27N+vWrbvlcZeGw4cP8+qrr3L48GGz5xs1asSCBQto164dUMHXcd25cycvvvgiQUFBODs7U716dXr27Gkx5hvg6NGjdOrUCRcXFzw8POjduzcXL14stN4vvviCoKAg7O3tqV27Nh988AE5OTl3dkRCCCGEEEKI/6xjx46pSatOp2Ps2LH89ddfLF++nPr16wO5Q4gnTpxYZB379+9XH48YMUJNWgE0Gg1jxowpdNv8FEVh+/btAPj6+vLGG2+YlQ8ePJhq1aqpPz/++OMWdXTu3Fl9HBsbC0BmtpHDl5I4EZOCwWgq8hjuxMGDB2nXrp1F0gq5Q6c7d+6sHlNZu61B0osWLSIxMZExY8bQoEEDrl27xqeffsrDDz/M1q1b6dixIwBnzpyhffv2NGvWjDVr1qDX63n//fdp27Ytx48fN7vZfObMmUyePJl33nmH4OBgDh06xKRJk7h8+TKLFy8u3aMVQgghhBBC3Ne++uor9fGcOXMYN26c+nPPnj2pU6cOCQkJrF+/nvj4eHx8fCzqMJluJoQODg4W5fb29oVum5+iKOj1egCqV69ucX+qRqOhSpUqakJ68uRJizpOnjyJztUb++r10TfoQs8FfxMWm4bBlDto1tlOx4O1PGld25OHH/CkcXUP7Gzu7Ipn3pXkjIwMABo0bMiAF1/F0x6++eYbjh49Sk5ODsOGDePixYvodLo7ep07dVtDhQt7Y2/cuEFgYCCNGjVSs+9+/frx119/ER4ejpubGwCRkZHUqVOHN998kzlz5gCQmJiIn58fQ4YM4euvv1brnDVrFpMmTSI0NNRi7LkMFRbWzGAwoCgKGo3mP31zvbAuErfCGkncCmskcVs6WrVqpV4xvH79Oi4uLmblo0eP5osvvgBy70Ht1KmTRR0HDx6kdevWADRr1oxDhw6ZvSfjxo1j7ty5AMyYMcNiRuA8tWvX5tKlS2i1Ws6fP282NDk+Pp4aNWqQlZUFgFarZfPmzTRo1ZY/T13lQHg8O0IiMNq5WtSr0aeBzhbF1tHseQdbLS1qVKJ1bS8equ1JgI8z3s72aLXFT0a1fedfPDX4VRz8GuDT8BGcazYh22jixJRgUEx07tyZXbt2AbBu3Tp69epVpnnZbf1GFPZthIuLCw0aNFBnvTIYDPz2228MGTJETVoBatasSYcOHdiwYYOauP7xxx/o9XqGDRtmVuewYcN477332LhxY7E3TQthTcLCwtRp7ps2bVrezRGiRCRuhTWSuBXW6H6J25iYGL755htOnDiBVqvl4YcfZujQoXh5eanbpKWlsXHjRmJiYnB3d6dbt27Url27VF6/4ARCBRmNxkK3za9Vq1Y0bdqUkJAQjh8/TpcuXRg3bhyVKlXihx9+4MsvvwRyhyIXzGXyGzZsGFOmTMFkMtGrVy++/PJLHn30UQ4fPszo0aPVpBVA5+nHc1/8iUvDbND8f/Jn54piMmJKjKKanZ7Q3ZvIunwaY9o10GixrVwTB/9GNOn0LBnO1UhKz+af8ET+CU9U67XVafB1c6CauyNVPRyo6u5Itf//30an4VhkMgcvJXHo4g2qDv4UgBwgJTMHB1stlxIzCPRx4c0331QT13379tGrV69bvxGl7K6/yklNTeXo0aPqMOHw8HAyMzNp0qSJxbZNmjRh27Zt6PV6HBwcCA0NBaBx48Zm21WtWhVvb2+1XAghhBBCCFGxKYrCtGnTmD59ully+PPPPzNp0iQ+//xzhg0bxgcffMDcuXNJT09Xt9FoNPTu3Zuvv/7aLMG9Ey1btuTQoUMAfPnll7zzzjtqWWJiIqtWrQLA1ta20Jwlrz2ff/45nTp1Iicnh507d7Jz506L7aZMmWJ2n2pBr732GosXL+by5cucPHmStm3bWmzjXfdBfDoMJt0jUH1OHxlC5qXjZF0+g92NOJ5o9xgbN25Uy3U6HSaTiZz4CHLiI9h3ZBNz5nzEMyNfYX9EEgcjkjgamUxcaiY5RoWY5ExikjOLOXNajBmpZMWcYsATrRjW43EaVnNXhx7nvyhZHvMR3XXi+vrrr5Oenq5eHk9MzM3uC5sS2tPTE0VRSE5OpmrVqiQmJmJvb292s3P+bfPqKk5YWBg1a9Y0O5lZWVmcOXMGgEqVKlGjRg2zfc6fP6+O3y74jVZCQgKXL18GoEaNGlSqVEktMxqNakLt6upqMRNZRESEutZTw4YNzYYUpKSkEBkZCUC1atUsFpY+ceIEiqLg6OhI3bp1zcqio6NJSkoCoF69emZj7W/cuKEuZuzj40PVqlXN9j116pT67V3BK9hxcXHEx8cDuTOm5R9KodfrOXv2LJD7fvj7+5vte+7cOTIzM9FoNBa/9NeuXVPH69esWdNsSneDwUBYWBiQ+wtQ8Nu1ixcvqsMOGjVqZDZ+Pjk5maioKCD3XgFvb2+zfUNCQgBwcnKiTp06ZmVRUVEkJycDqJOB5UlLS1OnGvf19aVKlSpm+4aFhWEwGLC3tycoKMisLDY2lmvXrgEQGBhoFs8ZGRmcP38eAC8vL9zc3DAYDGpcnD17Fr1ej06nU2euyxMfH09cXBwAtWrVwt3dXS3Lzs7m9OnTALi7u1OrVi2zfcPDw7lx4waQ+8VQ/iEbiYmJxMTEAODn52f2x8FkMqn3V7i4uBAQEGBW76VLl0hNTQVyp4W3s7NTy1JTU7l06RKQ++VTwREaoaGhGI1GHBwcqFevnllZTEyM+vtep04dnJyc1LL09HQuXLgAQOXKlS3+OJw5c4asrCxsbGxo2LChWdmVK1e4evUqkDtUR/qIO+sjjEYjGo0GR0fz4UjSR9xUWn1EwcXlpY/IdSd9RF5/m56eTkhIiPQR8jkCqPh9RF5s573v1tZHzJw5k6lTp1KYzMxMRowYwddff13oxD+KorBu3TpCQkLYsmULgYGBZuW300e88sorLFq0CID33nuPI0eO0KFDB65fv87SpUvV34U+ffqQmJhIbGxsoX1E3bp1mT9/PlOmTFHfozz29vaMHDmSHj16EBUVdcs+Yvv27Tz11FMWE9Xa+zfGp/3zaKs1JDeFV9BePsHlbd+QfTUcOzs7BgwYwJAh89Qldezs7JgwYQI9evQgMDCQRYsWqXnYzJkzePzxtjRxdKRrJz8qV25OjtFE/PUs4lIy2X/iHNcyDKRkg17jQFyqnvQsA039PAhwh8vH9/DhnNxJpyKNT9PstafRaDRqHzFv3jy17QXjrSzcVeI6efJkVq5cyRdffMGLine truncated
+      "image/png": "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 truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -1212,7 +1211,7 @@
    "outputs": [
     {
      "data": {
-      "image/png": "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 truncated
+      "image/png": "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 truncated
       "text/plain": [
        "<Figure size 1000x400 with 1 Axes>"
       ]
@@ -1534,7 +1533,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "3ba213859e934463a5aa61bd095dd71b",
+       "model_id": "594bd1b9610b440c97b7e2b9c44e549e",
        "version_major": 2,
        "version_minor": 0
       },

kf_book/gh_internal.py

@@ -172,19 +172,18 @@ def plot_estimate_chart_3():
         plt.ylim(157, 164.5)
 
 
-def plot_gh_results(weights, estimates, predictions, time_step=0):
-
+def plot_gh_results(weights, estimates, predictions, actual, time_step=0):
     n = len(weights)
     if time_step > 0:
         rng = range(1, n+1)
     else:
         rng = range(n, n+1)
     xs = range(n+1)
-    pred, = book_plots.plot_track(xs[1:], predictions, c='r', marker='v')
-    scale, = book_plots.plot_measurements(xs[1:], weights, color='k', lines=False)
-    est, = book_plots.plot_filter(xs, estimates, marker='o')
-
-    plt.legend([scale, est, pred], ['Measurement', 'Estimates', 'Predictions'], loc=4)
+    book_plots.plot_measurements(xs[1:], weights, color='k', lines=False)
+    book_plots.plot_filter(xs, estimates, marker='o', label='Estimates')
+    book_plots.plot_track(xs[1:], predictions, c='r', marker='v', label='Predictions')
+    plt.plot([xs[0], xs[-1]], actual, c='k', lw=1, label='Actual')
+    plt.legend(loc=4)
     book_plots.set_labels(x='day', y='weight (lbs)')
     plt.xlim([-1, n+1])
     plt.ylim([156.0, 173])