From 11aa42cbb3477a3976d4f29bf033a2fa1ad6557a Mon Sep 17 00:00:00 2001
From: Lawrence Anthony <35278035+larrymoralez@users.noreply.github.com>
Date: Mon, 8 Apr 2019 16:00:53 -0400
Subject: [PATCH] Created using Colaboratory

---
 ML_HW_2_4.ipynb | 234 ++++++++++++++++++++++++++++++++++--------------
 1 file changed, 167 insertions(+), 67 deletions(-)

diff --git a/ML_HW_2_4.ipynb b/ML_HW_2_4.ipynb
index ec91ed2..3430c1b 100644
--- a/ML_HW_2_4.ipynb
+++ b/ML_HW_2_4.ipynb
@@ -28,7 +28,11 @@
       "metadata": {
         "id": "IZZ8g2tvW1XS",
         "colab_type": "code",
-        "colab": {}
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 68
+        },
+        "outputId": "68f66d5d-6ee9-467f-82ae-e550a7facaac"
       },
       "cell_type": "code",
       "source": [
@@ -48,8 +52,24 @@
         "\n",
         "\n"
       ],
-      "execution_count": 0,
-      "outputs": []
+      "execution_count": 1,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "Using TensorFlow backend.\n"
+          ],
+          "name": "stderr"
+        },
+        {
+          "output_type": "stream",
+          "text": [
+            "Downloading data from https://www.cs.toronto.edu/~kriz/cifar-10-python.tar.gz\n",
+            "170500096/170498071 [==============================] - 30s 0us/step\n"
+          ],
+          "name": "stdout"
+        }
+      ]
     },
     {
       "metadata": {
@@ -70,18 +90,32 @@
       "metadata": {
         "id": "Zrv9LETKXCGD",
         "colab_type": "code",
-        "colab": {}
+        "colab": {
+          "base_uri": "https://localhost:8080/",
+          "height": 108
+        },
+        "outputId": "4b300e79-bb40-46a8-b070-134e950b314c"
       },
       "cell_type": "code",
       "source": [
         "#Set initial params\n",
         "batch_size = 32\n",
         "num_classes = 10\n",
-        "epochs = 10\n",
-        "RMS = keras.optimizers.rmsprop(lr=0.00001, decay=1e-6)"
+        "epochs = 40\n",
+        "RMS = keras.optimizers.rmsprop(lr=0.0001, decay=1e-6)"
       ],
-      "execution_count": 0,
-      "outputs": []
+      "execution_count": 3,
+      "outputs": [
+        {
+          "output_type": "stream",
+          "text": [
+            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/framework/op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.\n",
+            "Instructions for updating:\n",
+            "Colocations handled automatically by placer.\n"
+          ],
+          "name": "stdout"
+        }
+      ]
     },
     {
       "metadata": {
@@ -127,9 +161,9 @@
         "colab_type": "code",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 748
+          "height": 839
         },
-        "outputId": "49d20dfb-7508-4dc8-9bbd-4f678fe4f6a0"
+        "outputId": "1690802b-00dc-4f7f-a76b-14ddc836c6c1"
       },
       "cell_type": "code",
       "source": [
@@ -158,49 +192,52 @@
         "\n",
         "model.summary()"
       ],
-      "execution_count": 41,
+      "execution_count": 6,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
+            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3445: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.\n",
+            "Instructions for updating:\n",
+            "Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.\n",
             "_________________________________________________________________\n",
             "Layer (type)                 Output Shape              Param #   \n",
             "=================================================================\n",
-            "conv2d_13 (Conv2D)           (None, 32, 32, 32)        896       \n",
+            "conv2d_1 (Conv2D)            (None, 32, 32, 32)        896       \n",
             "_________________________________________________________________\n",
-            "activation_19 (Activation)   (None, 32, 32, 32)        0         \n",
+            "activation_1 (Activation)    (None, 32, 32, 32)        0         \n",
             "_________________________________________________________________\n",
-            "conv2d_14 (Conv2D)           (None, 30, 30, 32)        9248      \n",
+            "conv2d_2 (Conv2D)            (None, 30, 30, 32)        9248      \n",
             "_________________________________________________________________\n",
-            "activation_20 (Activation)   (None, 30, 30, 32)        0         \n",
+            "activation_2 (Activation)    (None, 30, 30, 32)        0         \n",
             "_________________________________________________________________\n",
-            "max_pooling2d_7 (MaxPooling2 (None, 15, 15, 32)        0         \n",
+            "max_pooling2d_1 (MaxPooling2 (None, 15, 15, 32)        0         \n",
             "_________________________________________________________________\n",
-            "dropout_10 (Dropout)         (None, 15, 15, 32)        0         \n",
+            "dropout_1 (Dropout)          (None, 15, 15, 32)        0         \n",
             "_________________________________________________________________\n",
-            "conv2d_15 (Conv2D)           (None, 15, 15, 64)        18496     \n",
+            "conv2d_3 (Conv2D)            (None, 15, 15, 64)        18496     \n",
             "_________________________________________________________________\n",
-            "activation_21 (Activation)   (None, 15, 15, 64)        0         \n",
+            "activation_3 (Activation)    (None, 15, 15, 64)        0         \n",
             "_________________________________________________________________\n",
-            "conv2d_16 (Conv2D)           (None, 13, 13, 64)        36928     \n",
+            "conv2d_4 (Conv2D)            (None, 13, 13, 64)        36928     \n",
             "_________________________________________________________________\n",
-            "activation_22 (Activation)   (None, 13, 13, 64)        0         \n",
+            "activation_4 (Activation)    (None, 13, 13, 64)        0         \n",
             "_________________________________________________________________\n",
-            "max_pooling2d_8 (MaxPooling2 (None, 6, 6, 64)          0         \n",
+            "max_pooling2d_2 (MaxPooling2 (None, 6, 6, 64)          0         \n",
             "_________________________________________________________________\n",
-            "dropout_11 (Dropout)         (None, 6, 6, 64)          0         \n",
+            "dropout_2 (Dropout)          (None, 6, 6, 64)          0         \n",
             "_________________________________________________________________\n",
-            "flatten_4 (Flatten)          (None, 2304)              0         \n",
+            "flatten_1 (Flatten)          (None, 2304)              0         \n",
             "_________________________________________________________________\n",
-            "dense_7 (Dense)              (None, 512)               1180160   \n",
+            "dense_1 (Dense)              (None, 512)               1180160   \n",
             "_________________________________________________________________\n",
-            "activation_23 (Activation)   (None, 512)               0         \n",
+            "activation_5 (Activation)    (None, 512)               0         \n",
             "_________________________________________________________________\n",
-            "dropout_12 (Dropout)         (None, 512)               0         \n",
+            "dropout_3 (Dropout)          (None, 512)               0         \n",
             "_________________________________________________________________\n",
-            "dense_8 (Dense)              (None, 10)                5130      \n",
+            "dense_2 (Dense)              (None, 10)                5130      \n",
             "_________________________________________________________________\n",
-            "activation_24 (Activation)   (None, 10)                0         \n",
+            "activation_6 (Activation)    (None, 10)                0         \n",
             "=================================================================\n",
             "Total params: 1,250,858\n",
             "Trainable params: 1,250,858\n",
@@ -220,7 +257,7 @@
       "cell_type": "code",
       "source": [
         "#Compile model\n",
-        "history = model.compile(loss='categorical_crossentropy',\n",
+        "model.compile(loss='categorical_crossentropy',\n",
         "              optimizer=RMS,\n",
         "              metrics=['accuracy'])"
       ],
@@ -231,11 +268,11 @@
       "metadata": {
         "id": "dZNamcq8XLDH",
         "colab_type": "code",
+        "outputId": "421705d9-52b2-4a81-ea99-170eabf47a0e",
         "colab": {
           "base_uri": "https://localhost:8080/",
-          "height": 357
-        },
-        "outputId": "3fd1c9ac-401d-4a44-cab4-7550a0023025"
+          "height": 1448
+        }
       },
       "cell_type": "code",
       "source": [
@@ -245,31 +282,94 @@
         "                              validation_data=(x_val, y_val), \n",
         "                              shuffle = True)"
       ],
-      "execution_count": 43,
+      "execution_count": 8,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
-            "Epoch 1/10\n",
-            "1250/1250 [==============================] - 250s 200ms/step - loss: 8.0570 - acc: 0.1212 - val_loss: 2.2076 - val_acc: 0.1808\n",
-            "Epoch 2/10\n",
-            "1250/1250 [==============================] - 251s 201ms/step - loss: 2.2998 - acc: 0.1644 - val_loss: 2.1074 - val_acc: 0.2318\n",
-            "Epoch 3/10\n",
-            "1250/1250 [==============================] - 250s 200ms/step - loss: 2.1498 - acc: 0.1977 - val_loss: 2.0315 - val_acc: 0.2638\n",
-            "Epoch 4/10\n",
-            "1250/1250 [==============================] - 250s 200ms/step - loss: 2.0869 - acc: 0.2159 - val_loss: 1.9731 - val_acc: 0.2760\n",
-            "Epoch 5/10\n",
-            "1250/1250 [==============================] - 251s 201ms/step - loss: 2.0410 - acc: 0.2374 - val_loss: 1.9423 - val_acc: 0.2881\n",
-            "Epoch 6/10\n",
-            "1250/1250 [==============================] - 251s 200ms/step - loss: 2.0109 - acc: 0.2507 - val_loss: 1.9169 - val_acc: 0.3021\n",
-            "Epoch 7/10\n",
-            "1250/1250 [==============================] - 251s 201ms/step - loss: 1.9848 - acc: 0.2631 - val_loss: 1.9003 - val_acc: 0.3086\n",
-            "Epoch 8/10\n",
-            "1250/1250 [==============================] - 251s 201ms/step - loss: 1.9650 - acc: 0.2729 - val_loss: 1.9024 - val_acc: 0.3022\n",
-            "Epoch 9/10\n",
-            "1250/1250 [==============================] - 249s 199ms/step - loss: 1.9426 - acc: 0.2822 - val_loss: 1.8730 - val_acc: 0.3151\n",
-            "Epoch 10/10\n",
-            "1250/1250 [==============================] - 250s 200ms/step - loss: 1.9339 - acc: 0.2868 - val_loss: 1.8517 - val_acc: 0.3200\n"
+            "WARNING:tensorflow:From /usr/local/lib/python3.6/dist-packages/tensorflow/python/ops/math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.\n",
+            "Instructions for updating:\n",
+            "Use tf.cast instead.\n",
+            "Epoch 1/40\n",
+            "1250/1250 [==============================] - 250s 200ms/step - loss: 3.9944 - acc: 0.2074 - val_loss: 1.9571 - val_acc: 0.2707\n",
+            "Epoch 2/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.9292 - acc: 0.2904 - val_loss: 1.8153 - val_acc: 0.3304\n",
+            "Epoch 3/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.8526 - acc: 0.3220 - val_loss: 1.7776 - val_acc: 0.3497\n",
+            "Epoch 4/40\n",
+            "1250/1250 [==============================] - 247s 197ms/step - loss: 1.8155 - acc: 0.3361 - val_loss: 1.8276 - val_acc: 0.3444\n",
+            "Epoch 5/40\n",
+            "1250/1250 [==============================] - 243s 195ms/step - loss: 1.7868 - acc: 0.3486 - val_loss: 1.7071 - val_acc: 0.3815\n",
+            "Epoch 6/40\n",
+            "1250/1250 [==============================] - 243s 195ms/step - loss: 1.7692 - acc: 0.3592 - val_loss: 1.7879 - val_acc: 0.3623\n",
+            "Epoch 7/40\n",
+            "1250/1250 [==============================] - 244s 195ms/step - loss: 1.7553 - acc: 0.3663 - val_loss: 1.7447 - val_acc: 0.3680\n",
+            "Epoch 8/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.7434 - acc: 0.3720 - val_loss: 1.7111 - val_acc: 0.3728\n",
+            "Epoch 9/40\n",
+            "1250/1250 [==============================] - 248s 198ms/step - loss: 1.7265 - acc: 0.3777 - val_loss: 1.7097 - val_acc: 0.3885\n",
+            "Epoch 10/40\n",
+            "1250/1250 [==============================] - 248s 199ms/step - loss: 1.7194 - acc: 0.3774 - val_loss: 1.6846 - val_acc: 0.3950\n",
+            "Epoch 11/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.7082 - acc: 0.3840 - val_loss: 1.6602 - val_acc: 0.4037\n",
+            "Epoch 12/40\n",
+            "1250/1250 [==============================] - 247s 198ms/step - loss: 1.6992 - acc: 0.3868 - val_loss: 1.8216 - val_acc: 0.3548\n",
+            "Epoch 13/40\n",
+            "1250/1250 [==============================] - 244s 195ms/step - loss: 1.6928 - acc: 0.3906 - val_loss: 1.8262 - val_acc: 0.3540\n",
+            "Epoch 14/40\n",
+            "1250/1250 [==============================] - 242s 194ms/step - loss: 1.6896 - acc: 0.3905 - val_loss: 1.7423 - val_acc: 0.3791\n",
+            "Epoch 15/40\n",
+            "1250/1250 [==============================] - 250s 200ms/step - loss: 1.6853 - acc: 0.3965 - val_loss: 1.7226 - val_acc: 0.3740\n",
+            "Epoch 16/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6778 - acc: 0.4011 - val_loss: 1.6710 - val_acc: 0.3997\n",
+            "Epoch 17/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6660 - acc: 0.4026 - val_loss: 1.6666 - val_acc: 0.3970\n",
+            "Epoch 18/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6657 - acc: 0.4011 - val_loss: 1.6077 - val_acc: 0.4217\n",
+            "Epoch 19/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6699 - acc: 0.3999 - val_loss: 1.6987 - val_acc: 0.3908\n",
+            "Epoch 20/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.6622 - acc: 0.4055 - val_loss: 1.6291 - val_acc: 0.4229\n",
+            "Epoch 21/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.6622 - acc: 0.4076 - val_loss: 1.7069 - val_acc: 0.3871\n",
+            "Epoch 22/40\n",
+            "1250/1250 [==============================] - 246s 196ms/step - loss: 1.6590 - acc: 0.4084 - val_loss: 1.6262 - val_acc: 0.4105\n",
+            "Epoch 23/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.6548 - acc: 0.4100 - val_loss: 1.6414 - val_acc: 0.4147\n",
+            "Epoch 24/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6543 - acc: 0.4108 - val_loss: 1.6497 - val_acc: 0.4149\n",
+            "Epoch 25/40\n",
+            "1250/1250 [==============================] - 248s 198ms/step - loss: 1.6506 - acc: 0.4119 - val_loss: 1.5792 - val_acc: 0.4313\n",
+            "Epoch 26/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.6490 - acc: 0.4117 - val_loss: 1.8788 - val_acc: 0.3828\n",
+            "Epoch 27/40\n",
+            "1250/1250 [==============================] - 250s 200ms/step - loss: 1.6459 - acc: 0.4122 - val_loss: 1.6605 - val_acc: 0.4152\n",
+            "Epoch 28/40\n",
+            "1250/1250 [==============================] - 247s 198ms/step - loss: 1.6380 - acc: 0.4160 - val_loss: 1.6252 - val_acc: 0.4079\n",
+            "Epoch 29/40\n",
+            "1250/1250 [==============================] - 246s 197ms/step - loss: 1.6401 - acc: 0.4143 - val_loss: 1.5580 - val_acc: 0.4373\n",
+            "Epoch 30/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6341 - acc: 0.4187 - val_loss: 1.7493 - val_acc: 0.4031\n",
+            "Epoch 31/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6293 - acc: 0.4232 - val_loss: 1.7227 - val_acc: 0.3948\n",
+            "Epoch 32/40\n",
+            "1250/1250 [==============================] - 244s 195ms/step - loss: 1.6254 - acc: 0.4225 - val_loss: 1.7047 - val_acc: 0.3981\n",
+            "Epoch 33/40\n",
+            "1250/1250 [==============================] - 245s 196ms/step - loss: 1.6240 - acc: 0.4222 - val_loss: 1.6199 - val_acc: 0.4206\n",
+            "Epoch 34/40\n",
+            "1250/1250 [==============================] - 255s 204ms/step - loss: 1.6226 - acc: 0.4254 - val_loss: 1.5965 - val_acc: 0.4305\n",
+            "Epoch 35/40\n",
+            "1250/1250 [==============================] - 251s 201ms/step - loss: 1.6209 - acc: 0.4234 - val_loss: 1.7353 - val_acc: 0.3845\n",
+            "Epoch 36/40\n",
+            "1250/1250 [==============================] - 252s 201ms/step - loss: 1.6205 - acc: 0.4199 - val_loss: 1.5984 - val_acc: 0.4280\n",
+            "Epoch 37/40\n",
+            "1250/1250 [==============================] - 250s 200ms/step - loss: 1.6236 - acc: 0.4240 - val_loss: 1.6065 - val_acc: 0.4278\n",
+            "Epoch 38/40\n",
+            "1250/1250 [==============================] - 249s 199ms/step - loss: 1.6166 - acc: 0.4270 - val_loss: 1.6688 - val_acc: 0.3969\n",
+            "Epoch 39/40\n",
+            "1250/1250 [==============================] - 248s 198ms/step - loss: 1.6150 - acc: 0.4254 - val_loss: 1.6253 - val_acc: 0.4177\n",
+            "Epoch 40/40\n",
+            "1250/1250 [==============================] - 247s 198ms/step - loss: 1.6112 - acc: 0.4286 - val_loss: 1.6302 - val_acc: 0.4224\n"
           ],
           "name": "stdout"
         }
@@ -283,7 +383,7 @@
           "base_uri": "https://localhost:8080/",
           "height": 376
         },
-        "outputId": "e3ba475b-4a54-4964-b85a-3fdd42b21558"
+        "outputId": "db1c2524-aca7-4d25-e913-93c04e175d10"
       },
       "cell_type": "code",
       "source": [
@@ -300,12 +400,12 @@
         "plt.legend()\n",
         "plt.show()"
       ],
-      "execution_count": 44,
+      "execution_count": 9,
       "outputs": [
         {
           "output_type": "display_data",
           "data": {
-            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAecAAAFnCAYAAACcvYGMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAe8AAAFnCAYAAACPasF4AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3XtAFWXi//HPgaO2KCogIF62NQ2V\niMyupkmSJNrFxbywlllauV5WTaoly0tp3lJDq03z0re01DIx97deylvfSnNTu6kYWt8tJTUQFBAU\nufz+cD2FDDgHGTiD79dfnufMzHmeM3I+M88884yjuLi4WAAAwDa8qrsCAADAPYQ3AAA2Q3gDAGAz\nhDcAADZDeAMAYDOENwAANkN447IzYcIExcTEKCYmRtdcc426dOniep2Tk+PWtmJiYpSenl7uMrNm\nzdKyZcsupcqV7uGHH9aqVatKlG3btk2dOnVSYWFhifKioiJ17txZ27ZtK3ebrVu31tGjR/Xxxx/r\nmWeeMf25Rt577z3Xv818x2atWrVKDz/8cKVsC6hOzuquAFDVnn/+ede/o6KiNGPGDN14440V2tb6\n9esvukx8fHyFtl3Vbr31VjmdTm3fvl2dOnVyle/YsUNeXl669dZbTW0nOjpa0dHRFa5HWlqaFi5c\nqL59+0oy9x0DlxvOvIELDBgwQC+//LK6d++u3bt3Kz09XYMHD1ZMTIyioqL05ptvupY9f7a5Y8cO\n9evXT7NmzVL37t0VFRWlf//735KkhIQE/eMf/5B07mBh+fLl6t27tzp16qRp06a5tjVv3jx16NBB\n999/v9555x1FRUUZ1u/9999X9+7dddddd+mBBx5QamqqpHNnlSNHjtTYsWPVrVs39ejRQwcOHJAk\nHTp0SH369FHXrl0VHx9f6uxakry8vNSzZ0+tWbOmRPmaNWvUs2dPeXl5lftdnPf7s9vyPnfTpk26\n99571a1bN/Xq1UvJycmSpLi4OP3yyy+KiYlRfn6+6zuWpLfffls9evRQTEyMhg4dqoyMDNd3PHfu\nXD3yyCPq0qWLHnnkEeXl5ZW1iw3t379fcXFxiomJUc+ePfXpp59Kkk6dOqXhw4ere/fuuvPOO/Xc\nc8/p7NmzZZYDVYHwBgzs2bNH//rXv9S+fXu9/vrratasmdavX6+33npLs2bN0pEjR0qts2/fPl13\n3XVat26d+vfvr9dff91w219++aVWrFihDz74QEuXLtXRo0d14MABLVy4UB9++KHefffdMs82jx8/\nrhdeeEFvvvmmPvroI/3xj390HRhI0v/+7/+qf//+2rBhg2655Ra99dZbkqSZM2eqQ4cO2rhxowYO\nHKjdu3cbbr9Xr17auHGjK/hOnz6tjz76SL169ZIk09/FeWV9bkFBgRISEjRp0iRt2LBBUVFRmj59\nuiRpypQpCgkJ0fr161W7dm3Xtr7++mstWrRIS5Ys0fr169WkSRPNmjXL9f769ev18ssv6+OPP1ZG\nRoY+/vjjMut1oaKiIo0ZM0YPPvig1q9fr8mTJys+Pl45OTlavXq16tevr3Xr1mnDhg3y9vbWwYMH\nyywHqgLhDRiIjIyUl9e5P4/nnntO48aNkyQ1b95cgYGBOnz4cKl16tatq65du0qSrrnmGv3yyy+G\n27733nvl7e2t4OBgBQQE6MiRI/ryyy918803KygoSHXq1NH9999vuG5AQIB27dqlxo0bS5JuvPFG\nHTp0yPV+y5YtFR4eLkkKCwtzBevOnTvVo0cPSVJERISuuuoqw+1feeWVat26tSv4Nm3apNDQUF15\n5ZVufRfnlfW5TqdT27ZtU7t27QzbYWTr1q3q1q2bAgICJEl9+vTR559/7no/MjJSDRs2lNPpVGho\naLkHFRc6fPiw0tPTdffdd0uSrr32WjVp0kTfffed/P399dVXX+mzzz5TUVGRnn/+ebVt27bMcqAq\ncM0bMNCgQQPXv7/77jvXGaaXl5fS0tJUVFRUah1fX1/Xv728vAyXkaR69eq5/u3t7a3CwkJlZWWV\n+Mzg4GDDdQsLCzV37lxt3rxZhYWFOnXqlFq0aGFYh/PblqSTJ0+W+Nz69euX2fZevXppzZo1uu++\n+7RmzRrXWbc738V55X3ukiVLlJSUpPz8fOXn58vhcJS5HUnKyMhQUFBQiW0dP378om03IyMjQ76+\nviXqUL9+fWVkZOjuu+/WyZMnNWfOHP3444+677779Mwzz6h79+6G5b/vLQCswpk3cBFPPfWUunXr\npg0bNmj9+vXy8/Or9M+oV6+ecnNzXa9//fVXw+XWrl2rzZs3a+nSpdqwYYNGjhxpavv169cvMZL+\n/LViI+ev9f/f//2fdu7cqe7du7vec/e7KOtzd+/erQULFuj111/Xhg0bNHny5Iu2oVGjRjpx4oTr\n9YkTJ9SoUaOLrmdGQECATp48qd8/p+nEiROus/y4uDi9//77Wrt2rfbu3avVq1eXWw5YjfAGLuL4\n8eMKDw+Xw+FQUlKS8vLySgRtZYiIiNCOHTuUkZGh/Pz8MkPg+PHjatq0qfz9/ZWZmal169bp1KlT\nF91+u3btXF3hu3fv1s8//1zmsvXq1VNUVJSef/55denSpcSZs7vfRVmfm5GRoYCAADVp0kR5eXlK\nSkpSbm6uiouL5XQ6lZubq4KCghLbuuOOO/Txxx8rMzNTkrR8+XJFRkZetO1mNGvWTI0bN9batWtd\ndU1PT1dERIRee+01rVy5UtK5HpFmzZrJ4XCUWQ5UBcIbuIhRo0Zp+PDhuvfee5Wbm6t+/fpp3Lhx\n5QaguyIiIhQbG6vY2Fg99NBD6tKli+Fy99xzj06cOKHo6GjFx8dr9OjROnr0aIlR60aeeuopbdmy\nRV27dtU777yj2267rdzle/Xqpe3bt5foMpfc/y7K+tzbb79dQUFB6tq1qwYNGqSBAwfK19dXI0eO\nVOvWrdWgQQN17NixxLiBiIgIPf7443rggQcUExOj7OxsPfHEE+W2w8jXX3/tuq8/JiZG/fv3l8Ph\n0OzZs7V06VJ1795dkydP1pw5c+Tj46OePXvqww8/VLdu3RQTE6NatWqpZ8+eZZYDVcHB87wBz1Bc\nXOw6c9u6dasSExPphgVgiDNvwANkZGTo1ltvVWpqqoqLi7Vu3TrXSGwAuBBn3oCHWLZsmRYvXiyH\nw6GrrrpKL774omvAFAD8HuENAIDN0G0OAIDNEN4AANiMbWZYS0vLNrWcn5+PMjMr9x7c6kR7PBvt\n8Wy0x7PRnosLDPQ1LK9xZ95Op3d1V6FS0R7PRns8G+3xbLSn4mpceAMAUNMR3gAA2AzhDQCAzRDe\nAADYDOENAIDNEN4AANgM4Q0AgM1YGt6nT59W165dtWrVqhLl27ZtU+/evdWvXz+99tprVlYBAOCB\nXnnlZQ0YMED9+9+vXr3u1ogRj2vs2KdMrbt27T/1ySdbynx/zpxZ+uWX1ArXbcSIx/XjjwcrvH5V\nsHSGtddff10NGjQoVT558mQtWrRIwcHBevDBB9WtWze1atXKyqq4JCU5lZhYWykpXgoNLdLo0fmK\njS2oks8GALuq7N/Ov/3tCQUG+uqtt97Vjz/+oBEjRptet0ePe8t9f9So+ArXyy4sC+8ffvhBBw8e\n1B133FGi/NChQ2rQoIFCQkIkSZGRkdq+fXuVhHdSklNDhvzB9To52fu/r/MIcAAoQ1X+du7evVPL\nly9Vbm6uRox4Ql99tUtbt25SUVGROnToqEGDHteiRfPVsGFDtWjRUqtWvSeHw0s//fR/uuOOOzVo\n0OMaMeJxjRnztLZs2aRTp3L0888/KTX1sEaOjFeHDh21dOn/aOPGj9SkSVMVFBQoLu4BtW9/Y6m6\n5OTk6MUXJyonJ1sFBQUaPfoptW7dRomJL2n//mQVFhYqNra3evS4V4mJL+mHH1J0+nS+q8xKlnWb\nT58+XQkJCaXK09LS5O/v73rt7++vtLQ0q6pRQmJibcPyOXOMywEAVf/b+cMPBzV79qtq06atJOkf\n/1ioN974H61b9/906lROiWX37durZ5+dqHnz3tQHH6wota1ffz2mmTPnatSoJ7VmzSplZZ3UqlXv\na/78xXryyQR9/fXuMuvx/vvLdM014XrllfkaNSper7wyW1lZJ7Vt22eaN2+xXn99kQoKClxly5cv\nd5VZzZIz79WrV6tdu3Zq3rx5pW3Tz8/H9LyxZU3knpJivHxKineZ63gCT65bRdAez0Z7PFt1tMfK\n305f3yvk41PbtZ2GDX0UFtZWTZsGSJIaNWqgJ54YKqfTqZMnT8jpLFTdunVUr94VatjQR9deG67m\nzQMlSQ6HQ4GBvqpd2yk/v7qqW7eOOnS4RYGBvmrduoXOnMnTqVMZatOmtZo1C5QUqOuui1DDhj4l\n2nF+/R9/TNHQoUMVGOirwMBbNHbsYbVs2UwtW16l8eOfVkxMjAYMiFPt2rXVsuVVGjp0aIkyK1kS\n3lu3btWhQ4e0detWHT16VLVr11bjxo112223KSgoSOnp6a5ljx07pqCgoItu0+yTWgIDfct8Allo\nqI+Sk0sfAISGFiotzTOfbFNee+yI9ng22uPZqqs9Vv12Bgb6Kjv7tHJz813tOnEiV8XFDqWlZevo\n0SNatGixFi9+Rz4+PhowoK8yMk7p1KkzqlXrtE6cyFVhYbFr3eLic//Ozy9QZub55f6gtLRsZWae\nUn5+gTIyTqmgoMi1ztmzhTpxIrfE93p+/bNnC5WRccr1XkFBodLSsjV16sv6/vv9+vjj9Xr//Q/0\n8suvaerUl/Xrrz/rvfdWucoqQ5U+VSwxMVEffPCB3nvvPfXp00fDhg3TbbfdJklq1qyZcnJydPjw\nYRUUFGjLli3q2LGjFdUoZfTofMPyUaOMywEA1ffbeeLECfn5+cnHx0fff79fR48e1dmzZy9pmyEh\nIfrxxx9UUFCgzMxM7d+fXOaybdqE6auvdkqS9uz5Ti1atNSRI7/o/feXq3XrNhoxYrROnjzpKrvm\nmmtcZVarsud5r1q1Sr6+voqOjtbEiRMVH39uNGCPHj3UokWLKqnDuYEVeZoz57cRk6NGMdocAMpT\nXb+dV18dqj/8wUdDhw7Stde2U8+evTRr1nRFRFxX4W36+wcoOjpGjz32kK68soXCwq6Rt7fxJdm+\nff+iKVOe18iRf1VRUZHGjPm7GjUK1J4932jTpo9Uq1Yt3X33fa6yuLg4SV66++77Klw/sxzFxcXF\nln9KJTDbVUQ3mWejPZ6N9ng22lM51q79p6KjY+Tt7a2HHorT7NmvKCgo+JK3a0V7yuo2r7IzbwAA\nPMHx48f1+OMDVatWbd11V0ylBHdVI7wBAJeVAQMe1oABD1d3NS4Jc5sDAGAzhDcAADZDeAMAYDOE\nNwAANkN4AwCq3JAhj2jPnj0lyubNe1XLli01XH737p167rmnJUkJCWNKvf/BByu0aNH8Mj/v4MED\n+vnnnyRJEyY8ozNnTle06urd+17l5lbvrJyENwDgouokrZRfZAc1CvGTX2QH1UlaeUnbi47upnXr\n1pUo27p1s7p2veui606bNtvtz/vkk806dOhnSdLzz09VnTpXuL0NT8KtYgCActVJWqn6Qwa5XjuT\n96r+kEHKknQmtneFtnnnnXdpxIjH9PDDf5Uk7d+frMDAQAUGBunLL3do4cJ5qlWrlnx9ffXCC9NK\nrHv33XfqX//apJ07/625c2fJ3z9AAQGNXI/4fPHFiUpL+1V5eXkaNOhxNW4cog8/XKVPPtksPz8/\njR//jN5+e4VycrI1deoLOnv2rLy8vJSQME4Oh0MvvjhRTZo01cGDBxQa2loJCeMM2/Drr8dKrD9j\nxjQ5nfX0wgvjdPx4uvLz8zV48BDdeOPNpcpuvfW2Cn1v53HmDQAol0/iLOPyOe6fAZ/n5+ev5s2b\na9++c13nmzd/rOjoGElSdna2JkyYrFdffUM+PnW1Y8d2w23Mn/+qxo2bpMTEf+jkyRP/XTdLN998\nq1599Q298MJULVo0Xy1bttItt3TQkCEjFBYW7lp/4cJ5uueennr11TcUG9tbixe/IUn6/vtkDRky\nXAsXvq3t2z9XdrbxrGkXrv/qq6/qhx8O6uTJE3rttQWaPftVZWVlGZZdKsIbAFAu75T9bpWbdc89\n92jTpo8lSZ9//r+64447JUkNGzbU9OmTNWLE4/rqq13KyjJ+0MeRI0d09dWhkqR27dpLknx96ys5\nea+GDh2kF1+cWOa60rmQvv76GyRJ7dvfqAMHvpckNW3aXAEBjeTl5aVGjQJLPUO8rPX37dunK6/8\nk3JzT2nSpHHavftLde16l2HZpSK8AQDlKgxt41a5WdHR0dq27VPt379PzZv/UfXr15ckTZ06SU88\n8bReffUNderUucz1vbx+i7Dzj+n4+OP1ysrK0muvLdSUKTMvUgOHa72zZwvkcJzb3oUPKin7ESAl\n1/fy8tIVV1yh+fP/R/fdd7+2b/9c06ZNMiy7VIQ3AKBcuaPjjctHlR717Y569eqpZcur9fbbb7q6\nzCXp1KkcBQc3VnZ2tnbv3lXmY0AbNQrUzz//R8XFxfrqq12Szj1GNCSkiby8vPTJJ5td6zocDhUW\nFpZYv23bMO3efe6Rn19/vUtt2rR1q/4Xrh8eHu56zvd117XTk08+o//85/8Myy4VA9YAAOU6E9tb\nWTp3jds7Zb8KQ9sod9SYCg9W+73o6BhNnjxBEyb8djbaq1cfDR06WM2b/1EPPPCQFi9+Q48/PqzU\nuo8/PkzPPfd3NW4c4nq4yB13RCkhYYz27duju+++T0FBQXrzzQW67rrrlZj4knx8fFzrP/roXzV1\n6iT985+r5XTW0jPPjFNBgfnHnF64/syZ05WTU6D581/Thx+ukpeXl/r3H6CQkCalyi4VjwT1cLTH\ns9Eez0Z7PBvtMbdNI3SbAwBgM4Q3AAA2Q3gDAGAzhDcAADZDeAMAYDOENwAANkN4AwBgM4Q3AAA2\nQ3gDAGAzhDcAADZDeAMAYDOENwAANkN4AwBgM4Q3AAA2Q3gDAGAzhDcAADZDeAMAYDOENwAANkN4\nAwBgM4Q3AAA2Q3gDAGAzhDcAADZDeAMAYDOENwAANkN4AwBgM4Q3AAA2Q3gDAGAzhDcAADZDeAMA\nYDOENwAANkN4AwBgM4Q3AAA247Rqw3l5eUpISNDx48d15swZDRs2TF26dHG9HxUVpcaNG8vb21uS\nNHPmTAUHB1tVHQAAagzLwnvLli0KDw/XY489ptTUVA0aNKhEeEvSggULVLduXauqAABAjWRZePfo\n0cP17yNHjnBWDQBAJbEsvM+Li4vT0aNHNW/evFLvTZgwQampqbrhhhsUHx8vh8NhdXUAALA9R3Fx\ncbHVH5KcnKynn35aa9ascQX06tWrdfvtt6tBgwYaPny4YmNjFRMTU+Y2CgoK5XR6W11VAAA8nmVn\n3nv27FFAQIBCQkLUtm1bFRYWKiMjQwEBAZKkP//5z65lO3furJSUlHLDOzMz19TnBgb6Ki0t+9Iq\n70Foj2ejPZ6N9ng22mNum0Ysu1Vs586dWrx4sSQpPT1dubm58vPzkyRlZ2dr8ODBys/PlyR9+eWX\nuvrqq62qCgAANYplZ95xcXF69tln1b9/f50+fVrjx4/X6tWr5evrq+joaHXu3Fn9+vVTnTp1FBYW\nVu5ZNwAA+I1l4X3FFVdo1qxZZb4/cOBADRw40KqPBwCgxmKGNQAAbIbwBgDAZghvAABshvAGAMBm\nCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghv\nAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAA\nbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG\n8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAG\nAMBmCG8AAGzGadWG8/LylJCQoOPHj+vMmTMaNmyYunTp4np/27Ztmj17try9vdW5c2cNHz7cqqoA\nAFCjWBbeW7ZsUXh4uB577DGlpqZq0KBBJcJ78uTJWrRokYKDg/Xggw+qW7duatWqlVXVAQCgxrAs\nvHv06OH695EjRxQcHOx6fejQITVo0EAhISGSpMjISG3fvp3wBgDABMvC+7y4uDgdPXpU8+bNc5Wl\npaXJ39/f9drf31+HDh0qdzt+fj5yOr1NfWZgoG/FKuuhaI9noz2ejfZ4NtpTMZaH9/Lly5WcnKyn\nnnpKa9askcPhqNB2MjNzTS0XGOirtLTsCn2GJ6I9no32eDba49loj7ltGrFstPmePXt05MgRSVLb\ntm1VWFiojIwMSVJQUJDS09Ndyx47dkxBQUFWVQUAgBrFsvDeuXOnFi9eLElKT09Xbm6u/Pz8JEnN\nmjVTTk6ODh8+rIKCAm3ZskUdO3a0qioAANQolnWbx8XF6dlnn1X//v11+vRpjR8/XqtXr5avr6+i\no6M1ceJExcfHSzo3uK1FixZWVQUAgBrFsvC+4oorNGvWrDLfv+mmm7RixQqrPh4AgBqLGdYAALAZ\nwhsAAJshvAEAsBnCGwAAmyG8AQCwGcIbAACbIbwBALAZwhsAAJshvAEAsBnCGwAAmzEV3nv27NGW\nLVskSS+//LIGDhyonTt3WloxAABgzFR4T548WS1atNDOnTv13Xffady4cZo7d67VdQMAAAZMhXed\nOnX0pz/9SZs2bVLfvn3VqlUreXnR4w4AQHUwlcB5eXlat26dNm7cqE6dOunEiRPKysqyum4AAMCA\nqfAeM2aM/vnPf+qJJ55QvXr1tGTJEj388MMWVw0AABgx9TzvW2+9VeHh4apXr57S09PVoUMHtW/f\n3uq6AQAAA6bOvCdNmqR169bpxIkTiouL09KlSzVx4kSLqwYAAIyYCu99+/apT58+WrdunWJjY5WY\nmKiffvrJ6roBAAADpsK7uLhYkrR161ZFRUVJkvLz862rFQAAKJOp8G7RooV69OihU6dOqW3btlq9\nerUaNGhgdd0AAIABUwPWJk+erJSUFLVs2VKS1KpVK82YMcPSigEAAGOmwvv06dPavHmz5syZI4fD\noXbt2qlVq1ZW1w0AABgw1W0+btw45eTkKC4uTn379lV6erqee+45q+sGAAAMmDrzTk9P1+zZs12v\nu3TpogEDBlhWKQAAUDbT06Pm5eW5Xufm5urMmTOWVQoAAJTN1Jl3v3791L17d4WHh0uS9u7dq1Gj\nRllaMQAAYMxUePfu3VsdO3bU3r175XA4NG7cOC1ZssTqugEAAAOmwluSQkJCFBIS4nr97bffWlIh\nAABQvgo/lPv8rGsAAKBqVTi8HQ5HZdYDAACYVG63eWRkpGFIFxcXKzMz07JKAQCAspUb3u+++25V\n1QMAAJhUbng3bdq0quoBAABMqvA1bwAAUD0IbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYA\nwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbKbcR4JeqhkzZmjXrl0qKCjQ\nkCFDdNddd7nei4qKUuPGjeXt7S1JmjlzpoKDg62sDgAANYJl4f3FF1/owIEDWrFihTIzMxUbG1si\nvCVpwYIFqlu3rlVVAACgRrIsvG+66SZFRERIkurXr6+8vDwVFha6zrQBAEDFOIqLi4ut/pAVK1Zo\n586deumll1xlUVFRat++vVJTU3XDDTcoPj5eDoejzG0UFBTK6ST4AQCw9Jq3JG3cuFErV67U4sWL\nS5SPHDlSt99+uxo0aKDhw4drw4YNiomJKXM7mZm5pj4vMNBXaWnZl1RnT0J7PBvt8Wy0x7PRHnPb\nNGLpaPNPP/1U8+bN04IFC+TrW7ICf/7znxUQECCn06nOnTsrJSXFyqoAAFBjWBbe2dnZmjFjhubP\nn6+GDRuWem/w4MHKz8+XJH355Ze6+uqrraoKAAA1imXd5mvXrlVmZqZGjx7tKrvlllvUunVrRUdH\nq3PnzurXr5/q1KmjsLCwcrvMAQDAbywL7379+qlfv35lvj9w4EANHDjQqo8HAKDGYoY1AABshvAG\nAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDA\nZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYI\nbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8A\nAGyG8AYAwGYIbwAAbIbwBgDAZghvAABshvAGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABs\nhvAGAMBmCG8AAGyG8AYAwGacVm58xowZ2rVrlwoKCjRkyBDdddddrve2bdum2bNny9vbW507d9bw\n4cOtrAoAADWGZeH9xRdf6MCBA1qxYoUyMzMVGxtbIrwnT56sRYsWKTg4WA8++KC6deumVq1aWVUd\nAABqDMvC+6abblJERIQkqX79+srLy1NhYaG8vb116NAhNWjQQCEhIZKkyMhIbd++nfAGAMAEy655\ne3t7y8fHR5K0cuVKde7cWd7e3pKktLQ0+fv7u5b19/dXWlqaVVUBAKBGsfSatyRt3LhRK1eu1OLF\niy9pO35+PnI6vU0tGxjoe0mf5Wloj2ejPZ6N9ng22lMxlob3p59+qnnz5mnhwoXy9f2tQUFBQUpP\nT3e9PnbsmIKCgsrdVmZmrqnPDAz0VVpadsUq7IFoj2ejPZ6N9ng22mNum0Ys6zbPzs7WjBkzNH/+\nfDVs2LDEe82aNVNOTo4OHz6sgoICbdmyRR07drSqKiXUSVopv8gOahTiJ7/IDqqTtLJKPhcAgMpi\n2Zn32rVrlZmZqdGjR7vKbrnlFrVu3VrR0dGaOHGi4uPjJUk9evRQixYtrKqKS52klao/ZJDrtTN5\nr+oPGaQsSWdie1v++QAAVAaBR3svAAAVSklEQVRHcXFxcXVXwgyzXRHldVv4RXaQM3lvqfKCsHBl\nbt12SfWzCt1Kno32eDba49loj7ltGrmsZljzTtnvVnlSklORkT4KCamnyEgfJSVZPr4PAICLuqzC\nuzC0jenypCSnhgz5g5KTvVVY6FBysreGDPkDAQ4AqHaXVXjnjo43Lh81plRZYmJtw2XnzDEuBwCg\nqlxW4X0mtrey5i9WQVi4ip1OFYSFK2v+YsPBaikpxl9NWeUAAFSVy64P+Exsb1Mjy0NDi5ScXHpS\nmNDQIiuqBQCAaZxGlmH06HzD8lGjjMsB2BNzP8COLrszb7NiYwsk5WnOnNpKSfFSaGiRRo3K/285\ngJqAuR9gV4R3OWJjCwhroAbzSZxlXD5nNuENj0a3eSXgfnDAntyd+wHwFIT3JeJ+cMC+3Jn7AfAk\nhPcl4n5wwL7cmfsB8CSE9yVy935wRrZWPb5zlMWduR8AT0Lf7iVy535wRrZWrjpJK+WTOEveKftV\nGNpGuaPjS32PfOe4GLNzPwCehDPvcpg5Y3PnfvCzL8w2XPbspJcvraKXofOh7EzeK0dhoSuUL9xH\n5Y0mBgC7IrzLYDYcYmMLNH9+nsLCCuV0FissrFDz5+cZ3mLWIDXZ8LPqHzYuR9nMhjKjiQHURIR3\nGdw5Y4uNLdDWrbn65Zccbd2aW+a94fsUZlieXEY5ymY2lBlNDKAmIrzLYMUZ21tNEgzL327691Jl\n5+8ddzp10XvHL8cBWWZDmdHEAGoiwrsM7pyxmQ3P1hNiFadl+kYROiunvlGE4rRMoeNjSyxX8t5x\nlXvvuNnu/ZrGbCgzmhhATUR4l8FsOLgTnrGxBYqa31P9w3bLx5mv/mG7FTW/Z6ludnfuHb9cB2S5\nE8pnYnsrc+s2pf+Socyt2whuALbHrWJlOBPbW1k6F4KuW5FGjSn1w+/u3Mhm5kt3595xr++Nu/G9\n9pcuT0pyKjHxtwetjB5t7wetcIsPgMsVZ97lMHPGZsW18bKeGW5UfqCW8WC3C8uTkpzaPORDLU++\nXqcLa2l58vXaPORDpnEFbOD8pTk5nZfNuBaUj/C+RFaMZnbn3vHn858xXPaFsyUHx33/fJKW6y+K\n0HdyqlAR+k7L9RelvJBUat3L+UErl+PgP3i231+aUw0Z18LByKUjvC+RFaOZS947rnLvHf+6TV/D\nQXBft+lbYrmBv0wz/KyHUqeXeO3ug1ZqUtBfroP/4Nlq2riWmngwUh0I70tk1Wjm8/eOnz2rcu8d\nHz06XysUp3b6RrV1Vu30jVYortRZepj2Ga7f9oJydwbLuRP07tz6Vl3c+ZHkzAFVxd1Lc57ee1TT\nDkaqC+FdCapzNLPZGd5ONm1ruH5Ws5LlKSle6qflJc7k+2m54WC5xMTahsteGPTu3Pp2fvnqOJs3\n+yPJmQOqkru3rXp67xGzHlYOwttDuXNmZ2aGt1rjjbvxa417osTrUcHLDK+NjwpeVmrddvvfM1y2\n3f73Sixn9dl8ZYW82R/Jy/3MwdPP7Goady7NVaT3qKr3I7MeVg7C2wNZcWZntnt/rGOq4frPOEpf\nM59Q23jZ8bVKLuvOrW9mg96KkDf7I2mnM4fK/oG2w5ldTfP7v11d5NJcRXqPqno/Muth5SC8PZBV\nZ3Zmuvf9jxo/JMWo/OqzxtfRLywPDS0y7F43uvXNbNBbEfJnYntr06NvaX+dc/XcXydCmx59q9T3\nZJczByt+oC/3Xofqcv5vV2fPlntpzg69R+4cjKBshLcHqs4zO3eCqai18bJFbUqWv9LxHcPu9bm3\nvVNqXbP3uFd2yEvngr7rwofU9sy5wX9tz3yjrgsfKhX0n3R82nCbn9z2lGF5dV3Dt+IH2k69Dpcj\nu/QemT0YQdkIbw9UnWd27nRpmV028vMZhstFbnupVJnZe9wrO+Ql80H/t88fMLw9b+S2B0qt687k\nON+OTVJG847yC/JTRvOO+nZs6Xvwz2/TzMGAFT/Qdul1uFyZvTx2Oe/HmjJmg/D2QNV5TcjdOcPN\nLOtOiMTGFmjjo2+X6Lre+OjbpQbhVXbIS+aDPiXFy/D2PKP1zU6O8+3YJN25cKBanzm3XOsz3+nO\nhQNLBbg7lwEyGhvfYWBUbvZWPnf+b5o9yHCnZ6I65xWwy4++mctjl+t155o0ZoPw9kDVfU3InVvf\nzCzr7q0uRiF24R+X2Vvk3JmtzmzQu3NAYHZynGZLZhou12xJya5vdy4DTCk2nn1vanHJ2ffcuZVv\nueIMex2WK66cbZZ9kOHuwEOzvRjuzCtg5oCgJv3oS5fv0/Zq0pgNR3FxcXF1V8KMtLRsU8sFBvqa\nXtYOakJ7zv/wXcjox8IvssO5UfYXKAgLP3eNrAKSkpyaM+e3B7KMGmX8QJbzQXKhCw8KzC4nSX5B\nfnKqsNSyZ+XUiV8z3F4uJKSeCgsdpZZzOov1yy85JcpCQuqpd+EKPaOpCtM+7VOYpuoZfeDsV2LZ\nyEgfJSd7l9pmWFihtm7NLVFmdll3lotIfl9jNcVVxykaq+/C+pT67Gnt/qlZv5S+NPFk03f096/u\ndb12Z/+YXdbr+tsUkLqn1HLHm12rot2fl9qm2QcAufuwoJrwe/B7Vd2eRiF+chSW/jsrdjqV/kuG\nwRrusaI9gYG+huWcecNyVtzq4g4z98GfX87M2bw709eanRznhzrGD5j58YJyd876Q0OLDLv3jcYF\nmJ2Yx51LC2aWMztXgGS+F8Od3gmzyzZINb4Lo/7hkuXu9iRYMUNhTZqyuLJZda2/OmZcJLxRJSr7\nVheruBP0ZqavNTs5zuEBTxoud3hAyWuT7lwGMLusOxPzuHNpwcztgWbnCpDMT/HrziBFs8vuk/HB\nVfIF5WZnHTy/rJFLmaHQivkPzA6kdGebVhyMmFnO3Wv9ZsY5VNeMi4Q3PEpNG0hj9tpixJRYbXr0\nLX3/34F63//3HvOIKbElljPbO+DOsu5MzGP2gMDs7YFm5wqQzPdiuNs7YWbZt5okGC73dtO/l3jt\nTk+CFbc7Vvb8B2YHUrqzTSsORswuZ3YuB8n8OIfquo5OeMOj1MSBNGYHAEZMiZX/oc904tcM+R/6\nrFRwnxen5fqm+DrlF9fSN8XXKU7Ly/xsMz0J7kzMY/aAwOztgWbnCpDM92JY0TvRekKs4UC90PEl\n95E7PQlW3O5Y2QcEZgdSurNNKw5G3HnOgpm5HCTp7AvG4Xt20sslXnt9b3xJz2u/tffME97wONX5\noBdPZ8WoZ3cvVZg5IDA7dsGdnhazB3bujElwZ5xD1Pye6h+2Wz7OfPUP262o+T1LLedOT4IVtztW\n9gFByzPG7bnKoNzs2Al3D0bMbNOK5yyYHedwoJbxJZWyyisL4Q3YiBVddFZcqjB7QOBuT4vZAzuz\nYxJ+v6zZcQ7lLedOT4IVtztW9gGB2YGUkvmxE+4cjJjdphXPWTA7zuH5fONbMl84a3yppbIQ3oCN\nWDEa34p5Bdw9o67snpbqet66uwdCZg4IrOhJMBvyZgdSSubHTrhzMGJ2m+48Z8GIUbnZcQ5ft+lr\neEnl6zZ9DdevLNzn7eFoj2er6vZYcR/871Vme+okrZTPnNnyTtmvwtA2yh01pkougbgzr4BVn29V\nuytz/5id/+DbsUlqtmSWrjqzTz/WCdPhAfGG4zHcuYf6t8/2VmhoYZmfbXabZv8u3J0DYPOQD0vN\nk3Dh5RJ3tlkRZd3nTXh7ONrj2aq6PVYHU03YP1Yf4FSmOkkr5ZM467egHx1f7n705P1Tke/9Yu0x\nu013/i7MHrS4s6zZg5GKILxtivZ4tupoj13O7KqL1bNoVZaKHIh58v6xoj3ubLO6enp+rypnWGPq\nHcBmzsT2ZgR+OQpD2xierXnaE7PKG3xox/17Jra3sqRKDVB3tnm5/V0Q3gBqlNzR8YZna5420U91\nP1PbClYE6OUWymYx2hxAjVLdT+Uzq7qnAoa9Ed4Aahyzc+lXp5o2FTCqlqXhnZKSoq5du2rp0qWl\n3ouKilL//v01YMAADRgwQMeOHbOyKgDgUWriVMCoOpZd887NzdWkSZPUoUOHMpdZsGCB6tata1UV\nAMCjcT0XFWXZmXft2rW1YMECBQUFWfURAABcliw783Y6nXI6y9/8hAkTlJqaqhtuuEHx8fFyOBxW\nVQcAgBqj2m4VGzlypG6//XY1aNBAw4cP14YNGxQTE1Pm8n5+PnI6vU1tu6yb2u2K9ng22uPZaI9n\noz0VU23h/ec//9n1786dOyslJaXc8M7MzDW1XU+egagiaI9noz2ejfZ4NtpjbptGquVWsezsbA0e\nPFj5+eeeIvPll1/q6quvro6qAABgO5adee/Zs0fTp09XamqqnE6nNmzYoKioKDVr1kzR0dHq3Lmz\n+vXrpzp16igsLKzcs24AAPAby8I7PDxcS5YsKfP9gQMHauDAgVZ9PAAANRYzrAEAYDOENwAANmOb\n53kDAIBzOPMGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABsptoeTGKFKVOm6JtvvpHD4dDY\nsWMVERFR3VWqsB07dmjUqFGuOd9DQ0M1bty4aq6V+1JSUjRs2DA9/PDDevDBB3XkyBE9/fTTKiws\nVGBgoF566SXVrl27uqtp2oXtSUhI0N69e9WwYUNJ0uDBg3XHHXdUbyXdMGPGDO3atUsFBQUaMmSI\nrr32Wlvvnwvbs3nzZtvun7y8PCUkJOj48eM6c+aMhg0bpjZt2th2/xi1Z8OGDbbdP+edPn1a99xz\nj4YNG6YOHTpU2f6pMeH973//Wz/99JNWrFihH374QWPHjtWKFSuqu1qX5Oabb9bcuXOruxoVlpub\nq0mTJqlDhw6usrlz56p///7q3r27Zs+erZUrV6p///7VWEvzjNojSWPGjFGXLl2qqVYV98UXX+jA\ngQNasWKFMjMzFRsbqw4dOth2/xi159Zbb7Xt/tmyZYvCw8P12GOPKTU1VYMGDVL79u1tu3+M2nP9\n9dfbdv+c9/rrr6tBgwaSqvb3rcZ0m2/fvl1du3aVJLVs2VInT55UTk5ONdfq8la7dm0tWLBAQUFB\nrrIdO3bozjvvlCR16dJF27dvr67quc2oPXZ20003ac6cOZKk+vXrKy8vz9b7x6g9hYWF1VyriuvR\no4cee+wxSdKRI0cUHBxs6/1j1B67++GHH3Tw4EFXb0FV7p8aE97p6eny8/Nzvfb391daWlo11ujS\nHTx4UH/961/1l7/8RZ9//nl1V8dtTqdTV1xxRYmyvLw8VzdSQECArfaRUXskaenSpXrooYf0xBNP\nKCMjoxpqVjHe3t7y8fGRJK1cuVKdO3e29f4xao+3t7dt9895cXFxevLJJzV27Fhb75/zft8eyb5/\nP5I0ffp0JSQkuF5X5f6pMd3mF7L7rK9/+tOfNGLECHXv3l2HDh3SQw89pI8++sg217fMsPs+kqSe\nPXuqYcOGatu2rd544w29+uqrGj9+fHVXyy0bN27UypUrtXjxYt11112ucrvun9+3Z8+ePbbfP8uX\nL1dycrKeeuqpEvvErvvn9+0ZO3asbffP6tWr1a5dOzVv3tzwfav3T4058w4KClJ6errr9a+//qrA\nwMBqrNGlCQ4OVo8ePeRwOPTHP/5RjRo10rFjx6q7WpfMx8dHp0+fliQdO3bM9l3QHTp0UNu2bSVJ\nUVFRSklJqeYauefTTz/VvHnztGDBAvn6+tp+/1zYHjvvnz179ujIkSOSpLZt26qwsFB169a17f4x\nak9oaKht98/WrVu1adMm9e3bV++//77+8Y9/VOnfT40J744dO2rDhg2SpL179yooKEj16tWr5lpV\n3Jo1a7Ro0SJJUlpamo4fP14jrhHddtttrv300Ucf6fbbb6/mGl2av/3tbzp06JCkc9e7zt8dYAfZ\n2dmaMWOG5s+f7xrta+f9Y9QeO++fnTt3avHixZLOXRbMzc219f4xas/48eNtu38SExP1wQcf6L33\n3lOfPn00bNiwKt0/NeqpYjNnztTOnTvlcDg0YcIEtWnTprqrVGE5OTl68sknlZWVpbNnz2rEiBGK\njIys7mq5Zc+ePZo+fbpSU1PldDoVHBysmTNnKiEhQWfOnFGTJk00depU1apVq7qraopRex588EG9\n8cYb+sMf/iAfHx9NnTpVAQEB1V1VU1asWKFXXnlFLVq0cJVNmzZNzz33nC33j1F7evXqpaVLl9py\n/5w+fVrPPvusjhw5otOnT2vEiBEKDw/X3//+d1vuH6P2+Pj46KWXXrLl/vm9V155RU2bNlWnTp2q\nbP/UqPAGAOByUGO6zQEAuFwQ3gAA2AzhDQCAzRDeAADYDOENAIDN1NgZ1gBIhw8fVkxMjK6//voS\n5ZGRkXr00Ucvefs7duxQYmKili1bdsnbAmAe4Q3UcP7+/lqyZEl1VwNAJSK8gctUWFiYhg0bph07\ndujUqVOaNm2aQkND9c0332jatGlyOp1yOBwaP368WrVqpf/85z8aN26cioqKVKdOHU2dOlWSVFRU\npAkTJig5OVm1a9fW/PnzJUnx8fHKyspSQUGBunTpoqFDh1Znc4EahWvewGWqsLBQV199tZYsWaK/\n/OUvrmfHP/3003rmmWe0ZMkSPfLII3r++eclSRMmTNDgwYP1zjvv6P7779e6desknXss4t/+9je9\n9957cjqd+uyzz7Rt2zYVFBTo3Xff1fLly+Xj46OioqJqaytQ03DmDdRwGRkZGjBgQImyp556SpLU\nqVMnSVL79u21aNEiZWVl6fjx44qIiJAk3XzzzRozZowk6dtvv9XNN98sSbr77rslnbvmfdVVV6lR\no0aSpMaNGysrK0tRUVGaO3euRo0apcjISPXp00deXpwrAJWF8AZquPKuef9+dmSHwyGHw1Hm+5IM\nz569vb1LlQUEBOjDDz/UV199pU2bNun+++9XUlKS4fPQAbiPQ2HgMvbFF19Iknbt2qXWrVvL19dX\ngYGB+uabbyRJ27dvV7t27SSdOzv/9NNPJUlr167V7Nmzy9zuZ599pq1bt+qGG27Q008/LR8fHx0/\nftzi1gCXD868gRrOqNu8WbNmkqR9+/Zp2bJlOnnypKZPny5Jmj59uqZNmyZvb295eXlp4sSJkqRx\n48Zp3Lhxevfdd+V0OjVlyhT9/PPPhp/ZokULJSQkaOHChfL29lanTp3UtGlT6xoJXGZ4qhhwmWrd\nurX27t0rp5NjeMBu6DYHAMBmOPMGAMBmOPMGAMBmCG8AAGyG8AYAwGYIbwAAbIbwBgDAZghvAABs\n5v8D3da0duNTWj4AAAAASUVORK5CYII=\n",
             "text/plain": [
               "<Figure size 576x396 with 1 Axes>"
             ]
@@ -324,7 +424,7 @@
           "base_uri": "https://localhost:8080/",
           "height": 376
         },
-        "outputId": "a2e2ef43-53bd-43c8-e6f3-0ec05a6ff0f1"
+        "outputId": "185a4caf-9c8e-4de8-cbd1-5f6e48d4d6fc"
       },
       "cell_type": "code",
       "source": [
@@ -339,12 +439,12 @@
         "plt.legend()\n",
         "plt.show()"
       ],
-      "execution_count": 45,
+      "execution_count": 10,
       "outputs": [
         {
           "output_type": "display_data",
           "data": {
-            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfsAAAFnCAYAAAChL+DqAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nU9iLiIgYnMJeRETE4P4/ztnYc0Uxp1oAAAAASUVORK5CYII=\n",
+            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfUAAAFnCAYAAAC/5tBZAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRETEJBTURURETOL/A+TTGjkFfCQUAAAAAElFTkSuQmCC\n",
             "text/plain": [
               "<Figure size 576x396 with 1 Axes>"
             ]
@@ -363,22 +463,22 @@
           "base_uri": "https://localhost:8080/",
           "height": 68
         },
-        "outputId": "9dbd18c6-84f7-44a1-e08f-2f5b23219131"
+        "outputId": "125812d7-ebbb-4b2b-8e19-5a2bfc92df02"
       },
       "cell_type": "code",
       "source": [
-        "test_score = model.evaluate(x_test, y_test, verbose=1)\n",
-        "print('Loss:', test_score[0])\n",
-        "print('Accuracy:', test_score[1])"
+        "scores = model.evaluate(x_test, y_test, verbose=1)\n",
+        "print('Loss:', scores[0])\n",
+        "print('Accuracy:', scores[1])"
       ],
-      "execution_count": 46,
+      "execution_count": 11,
       "outputs": [
         {
           "output_type": "stream",
           "text": [
             "10000/10000 [==============================] - 17s 2ms/step\n",
-            "Loss: 1.8413412532806397\n",
-            "Accuracy: 0.3311\n"
+            "Loss: 1.6131126216888427\n",
+            "Accuracy: 0.4167\n"
           ],
           "name": "stdout"
         }