diff --git a/Immobilienpreise.ipynb b/Immobilienpreise.ipynb index 101973cbc98bc4a5dc75725f2a0d152ecfdb5c83..30d79336192955036ba9881d88c4e731bebb5df7 100644 --- a/Immobilienpreise.ipynb +++ b/Immobilienpreise.ipynb @@ -7,263 +7,6 @@ "1. Importieren der Daten mit pandas" ] }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": {}, - "outputs": [], - "source": [ - "import pandas as pd\n", - "pd = pd.read_csv('housepricedata.csv')" - ] - }, - { - "cell_type": "code", - "execution_count": 2, - "metadata": {}, - "outputs": [ - { - "data": { - "text/html": [ - "<div>\n", - "<style scoped>\n", - " .dataframe tbody tr th:only-of-type {\n", - " vertical-align: middle;\n", - " }\n", - "\n", - " .dataframe tbody tr th {\n", - " vertical-align: top;\n", - " }\n", - "\n", - " .dataframe thead th {\n", - " text-align: right;\n", - " }\n", - "</style>\n", - "<table border=\"1\" class=\"dataframe\">\n", - " <thead>\n", - " <tr style=\"text-align: right;\">\n", - " <th></th>\n", - " <th>LotArea</th>\n", - " <th>OverallQual</th>\n", - " <th>OverallCond</th>\n", - " <th>TotalBsmtSF</th>\n", - " <th>FullBath</th>\n", - " <th>HalfBath</th>\n", - " <th>BedroomAbvGr</th>\n", - " <th>TotRmsAbvGrd</th>\n", - " <th>Fireplaces</th>\n", - " <th>GarageArea</th>\n", - " <th>AboveMedianPrice</th>\n", - " </tr>\n", - " </thead>\n", - " <tbody>\n", - " <tr>\n", - " <th>0</th>\n", - " <td>8450</td>\n", - " <td>7</td>\n", - " <td>5</td>\n", - " <td>856</td>\n", - " <td>2</td>\n", - " <td>1</td>\n", - " <td>3</td>\n", - " <td>8</td>\n", - " <td>0</td>\n", - " <td>548</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1</th>\n", - " <td>9600</td>\n", - " <td>6</td>\n", - " <td>8</td>\n", - " <td>1262</td>\n", - " <td>2</td>\n", - " <td>0</td>\n", - " <td>3</td>\n", - " <td>6</td>\n", - " <td>1</td>\n", - " <td>460</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>2</th>\n", - " <td>11250</td>\n", - " <td>7</td>\n", - " <td>5</td>\n", - " <td>920</td>\n", - " <td>2</td>\n", - " <td>1</td>\n", - " <td>3</td>\n", - " <td>6</td>\n", - " <td>1</td>\n", - " <td>608</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>3</th>\n", - " <td>9550</td>\n", - " <td>7</td>\n", - " <td>5</td>\n", - " <td>756</td>\n", - " <td>1</td>\n", - " <td>0</td>\n", - " <td>3</td>\n", - " <td>7</td>\n", - " <td>1</td>\n", - " <td>642</td>\n", - " <td>0</td>\n", - " </tr>\n", - " <tr>\n", - " <th>4</th>\n", - " <td>14260</td>\n", - " <td>8</td>\n", - " <td>5</td>\n", - " <td>1145</td>\n", - " <td>2</td>\n", - " <td>1</td>\n", - " <td>4</td>\n", - " <td>9</td>\n", - " <td>1</td>\n", - " <td>836</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>...</th>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " <td>...</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1455</th>\n", - " <td>7917</td>\n", - " <td>6</td>\n", - " <td>5</td>\n", - " <td>953</td>\n", - " <td>2</td>\n", - " <td>1</td>\n", - " <td>3</td>\n", - " <td>7</td>\n", - " <td>1</td>\n", - " <td>460</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1456</th>\n", - " <td>13175</td>\n", - " <td>6</td>\n", - " <td>6</td>\n", - " <td>1542</td>\n", - " <td>2</td>\n", - " <td>0</td>\n", - " <td>3</td>\n", - " <td>7</td>\n", - " <td>2</td>\n", - " <td>500</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1457</th>\n", - " <td>9042</td>\n", - " <td>7</td>\n", - " <td>9</td>\n", - " <td>1152</td>\n", - " <td>2</td>\n", - " <td>0</td>\n", - " <td>4</td>\n", - " <td>9</td>\n", - " <td>2</td>\n", - " <td>252</td>\n", - " <td>1</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1458</th>\n", - " <td>9717</td>\n", - " <td>5</td>\n", - " <td>6</td>\n", - " <td>1078</td>\n", - " <td>1</td>\n", - " <td>0</td>\n", - " <td>2</td>\n", - " <td>5</td>\n", - " <td>0</td>\n", - " <td>240</td>\n", - " <td>0</td>\n", - " </tr>\n", - " <tr>\n", - " <th>1459</th>\n", - " <td>9937</td>\n", - " <td>5</td>\n", - " <td>6</td>\n", - " <td>1256</td>\n", - " <td>1</td>\n", - " <td>1</td>\n", - " <td>3</td>\n", - " <td>6</td>\n", - " <td>0</td>\n", - " <td>276</td>\n", - " <td>0</td>\n", - " </tr>\n", - " </tbody>\n", - "</table>\n", - "<p>1460 rows × 11 columns</p>\n", - "</div>" - ], - "text/plain": [ - " LotArea OverallQual OverallCond TotalBsmtSF FullBath HalfBath \\\n", - "0 8450 7 5 856 2 1 \n", - "1 9600 6 8 1262 2 0 \n", - "2 11250 7 5 920 2 1 \n", - "3 9550 7 5 756 1 0 \n", - "4 14260 8 5 1145 2 1 \n", - "... ... ... ... ... ... ... \n", - "1455 7917 6 5 953 2 1 \n", - "1456 13175 6 6 1542 2 0 \n", - "1457 9042 7 9 1152 2 0 \n", - "1458 9717 5 6 1078 1 0 \n", - "1459 9937 5 6 1256 1 1 \n", - "\n", - " BedroomAbvGr TotRmsAbvGrd Fireplaces GarageArea AboveMedianPrice \n", - "0 3 8 0 548 1 \n", - "1 3 6 1 460 1 \n", - "2 3 6 1 608 1 \n", - "3 3 7 1 642 0 \n", - "4 4 9 1 836 1 \n", - "... ... ... ... ... ... \n", - "1455 3 7 1 460 1 \n", - "1456 3 7 2 500 1 \n", - "1457 4 9 2 252 1 \n", - "1458 2 5 0 240 0 \n", - "1459 3 6 0 276 0 \n", - "\n", - "[1460 rows x 11 columns]" - ] - }, - "execution_count": 2, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "pd" - ] - }, - { - "cell_type": "code", - "execution_count": 3, - "metadata": {}, - "outputs": [], - "source": [ - "datensatz = pd.values" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -271,17 +14,6 @@ "2. Aufteilen des Datensatz (Splitting)" ] }, - { - "cell_type": "code", - "execution_count": 4, - "metadata": {}, - "outputs": [], - "source": [ - "X = datensatz[:, 0:10]\n", - "\n", - "Y = datensatz[:, 10]" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -289,43 +21,6 @@ "3. Skalieren der X-Werte" ] }, - { - "cell_type": "code", - "execution_count": 5, - "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "array([[0.0334198 , 0.66666667, 0.5 , ..., 0.5 , 0. ,\n", - " 0.3864598 ],\n", - " [0.03879502, 0.55555556, 0.875 , ..., 0.33333333, 0.33333333,\n", - " 0.32440056],\n", - " [0.04650728, 0.66666667, 0.5 , ..., 0.33333333, 0.33333333,\n", - " 0.42877292],\n", - " ...,\n", - " [0.03618687, 0.66666667, 1. , ..., 0.58333333, 0.66666667,\n", - " 0.17771509],\n", - " [0.03934189, 0.44444444, 0.625 , ..., 0.25 , 0. ,\n", - " 0.16925247],\n", - " [0.04037019, 0.44444444, 0.625 , ..., 0.33333333, 0. ,\n", - " 0.19464034]])" - ] - }, - "execution_count": 5, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "from sklearn import preprocessing\n", - "\n", - "skalierer = preprocessing.MinMaxScaler()\n", - "X_skaliert = skalierer.fit_transform(X)\n", - "\n", - "X_skaliert" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -333,26 +28,6 @@ "4. Aufteilung in Trainings-, Validierungs- und Testdaten" ] }, - { - "cell_type": "code", - "execution_count": 6, - "metadata": {}, - "outputs": [], - "source": [ - "from sklearn.model_selection import train_test_split\n", - "\n", - "X_training, X_valid_und_test, Y_training, Y_valid_und_test = train_test_split(X_skaliert, Y, test_size=0.3)" - ] - }, - { - "cell_type": "code", - "execution_count": 7, - "metadata": {}, - "outputs": [], - "source": [ - "X_valid, X_test, Y_valid, Y_test = train_test_split(X_valid_und_test, Y_valid_und_test, test_size=0.5)" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -364,442 +39,6 @@ "- Inputschicht 10, Outputschicht 1, 32 Neuronen pro Schicht\n" ] }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/conda/lib/python3.11/site-packages/h5py/__init__.py:36: UserWarning: h5py is running against HDF5 1.14.3 when it was built against 1.14.2, this may cause problems\n", - " _warn((\"h5py is running against HDF5 {0} when it was built against {1}, \"\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Epoch 1/200\n", - "32/32 [==============================] - 1s 7ms/step - loss: 0.7008 - accuracy: 0.5010 - val_loss: 0.7011 - val_accuracy: 0.4475\n", - "Epoch 2/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6955 - accuracy: 0.5000 - val_loss: 0.6958 - val_accuracy: 0.4521\n", - "Epoch 3/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6919 - accuracy: 0.5098 - val_loss: 0.6923 - val_accuracy: 0.4977\n", - "Epoch 4/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6887 - accuracy: 0.6018 - val_loss: 0.6892 - val_accuracy: 0.5753\n", - "Epoch 5/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6858 - accuracy: 0.6409 - val_loss: 0.6861 - val_accuracy: 0.6027\n", - "Epoch 6/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6829 - accuracy: 0.6742 - val_loss: 0.6831 - val_accuracy: 0.6530\n", - "Epoch 7/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6798 - accuracy: 0.7006 - val_loss: 0.6798 - val_accuracy: 0.6804\n", - "Epoch 8/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6767 - accuracy: 0.7299 - val_loss: 0.6766 - val_accuracy: 0.7078\n", - "Epoch 9/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6736 - accuracy: 0.7495 - val_loss: 0.6733 - val_accuracy: 0.7443\n", - "Epoch 10/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6703 - accuracy: 0.7730 - val_loss: 0.6700 - val_accuracy: 0.7443\n", - "Epoch 11/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6669 - accuracy: 0.7798 - val_loss: 0.6665 - val_accuracy: 0.7443\n", - "Epoch 12/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6634 - accuracy: 0.7877 - val_loss: 0.6630 - val_accuracy: 0.7489\n", - "Epoch 13/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6597 - accuracy: 0.7886 - val_loss: 0.6591 - val_accuracy: 0.7671\n", - "Epoch 14/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6557 - accuracy: 0.7935 - val_loss: 0.6546 - val_accuracy: 0.7763\n", - "Epoch 15/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6512 - accuracy: 0.7935 - val_loss: 0.6498 - val_accuracy: 0.7808\n", - "Epoch 16/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6459 - accuracy: 0.7984 - val_loss: 0.6438 - val_accuracy: 0.7900\n", - "Epoch 17/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6398 - accuracy: 0.7994 - val_loss: 0.6373 - val_accuracy: 0.7945\n", - "Epoch 18/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6333 - accuracy: 0.8033 - val_loss: 0.6303 - val_accuracy: 0.8037\n", - "Epoch 19/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6267 - accuracy: 0.8072 - val_loss: 0.6236 - val_accuracy: 0.7991\n", - "Epoch 20/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6197 - accuracy: 0.8141 - val_loss: 0.6162 - val_accuracy: 0.8037\n", - "Epoch 21/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6121 - accuracy: 0.8200 - val_loss: 0.6082 - val_accuracy: 0.8037\n", - "Epoch 22/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.6040 - accuracy: 0.8249 - val_loss: 0.5999 - val_accuracy: 0.8219\n", - "Epoch 23/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5953 - accuracy: 0.8288 - val_loss: 0.5898 - val_accuracy: 0.7991\n", - "Epoch 24/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5858 - accuracy: 0.8307 - val_loss: 0.5796 - val_accuracy: 0.8037\n", - "Epoch 25/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5756 - accuracy: 0.8268 - val_loss: 0.5693 - val_accuracy: 0.8082\n", - "Epoch 26/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5649 - accuracy: 0.8415 - val_loss: 0.5585 - val_accuracy: 0.8265\n", - "Epoch 27/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5535 - accuracy: 0.8366 - val_loss: 0.5444 - val_accuracy: 0.8265\n", - "Epoch 28/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5417 - accuracy: 0.8395 - val_loss: 0.5342 - val_accuracy: 0.8311\n", - "Epoch 29/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5297 - accuracy: 0.8434 - val_loss: 0.5199 - val_accuracy: 0.8219\n", - "Epoch 30/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5170 - accuracy: 0.8434 - val_loss: 0.5073 - val_accuracy: 0.8219\n", - "Epoch 31/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.5039 - accuracy: 0.8513 - val_loss: 0.4917 - val_accuracy: 0.8356\n", - "Epoch 32/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4912 - accuracy: 0.8444 - val_loss: 0.4804 - val_accuracy: 0.8265\n", - "Epoch 33/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4778 - accuracy: 0.8483 - val_loss: 0.4637 - val_accuracy: 0.8402\n", - "Epoch 34/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4653 - accuracy: 0.8513 - val_loss: 0.4496 - val_accuracy: 0.8447\n", - "Epoch 35/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4527 - accuracy: 0.8523 - val_loss: 0.4370 - val_accuracy: 0.8356\n", - "Epoch 36/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4408 - accuracy: 0.8591 - val_loss: 0.4261 - val_accuracy: 0.8402\n", - "Epoch 37/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4290 - accuracy: 0.8483 - val_loss: 0.4097 - val_accuracy: 0.8539\n", - "Epoch 38/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.4187 - accuracy: 0.8542 - val_loss: 0.4030 - val_accuracy: 0.8402\n", - "Epoch 39/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.4089 - accuracy: 0.8562 - val_loss: 0.3891 - val_accuracy: 0.8539\n", - "Epoch 40/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.3986 - accuracy: 0.8601 - val_loss: 0.3778 - val_accuracy: 0.8493\n", - "Epoch 41/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3899 - accuracy: 0.8591 - val_loss: 0.3723 - val_accuracy: 0.8630\n", - "Epoch 42/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3826 - accuracy: 0.8630 - val_loss: 0.3609 - val_accuracy: 0.8721\n", - "Epoch 43/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3751 - accuracy: 0.8620 - val_loss: 0.3552 - val_accuracy: 0.8721\n", - "Epoch 44/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3680 - accuracy: 0.8620 - val_loss: 0.3404 - val_accuracy: 0.8813\n", - "Epoch 45/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3616 - accuracy: 0.8601 - val_loss: 0.3394 - val_accuracy: 0.8813\n", - "Epoch 46/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3558 - accuracy: 0.8630 - val_loss: 0.3283 - val_accuracy: 0.8813\n", - "Epoch 47/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3502 - accuracy: 0.8630 - val_loss: 0.3161 - val_accuracy: 0.8904\n", - "Epoch 48/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3468 - accuracy: 0.8669 - val_loss: 0.3107 - val_accuracy: 0.8950\n", - "Epoch 49/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3417 - accuracy: 0.8611 - val_loss: 0.3095 - val_accuracy: 0.8858\n", - "Epoch 50/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3380 - accuracy: 0.8679 - val_loss: 0.3041 - val_accuracy: 0.8950\n", - "Epoch 51/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3341 - accuracy: 0.8650 - val_loss: 0.2946 - val_accuracy: 0.8950\n", - "Epoch 52/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3305 - accuracy: 0.8669 - val_loss: 0.2918 - val_accuracy: 0.8950\n", - "Epoch 53/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3269 - accuracy: 0.8669 - val_loss: 0.2946 - val_accuracy: 0.8995\n", - "Epoch 54/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3250 - accuracy: 0.8630 - val_loss: 0.2848 - val_accuracy: 0.8995\n", - "Epoch 55/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.3236 - accuracy: 0.8659 - val_loss: 0.2825 - val_accuracy: 0.9087\n", - "Epoch 56/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3209 - accuracy: 0.8708 - val_loss: 0.2801 - val_accuracy: 0.9132\n", - "Epoch 57/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3176 - accuracy: 0.8640 - val_loss: 0.2687 - val_accuracy: 0.9041\n", - "Epoch 58/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3162 - accuracy: 0.8699 - val_loss: 0.2699 - val_accuracy: 0.9178\n", - "Epoch 59/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3127 - accuracy: 0.8620 - val_loss: 0.2682 - val_accuracy: 0.9224\n", - "Epoch 60/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3101 - accuracy: 0.8679 - val_loss: 0.2710 - val_accuracy: 0.9178\n", - "Epoch 61/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3099 - accuracy: 0.8699 - val_loss: 0.2612 - val_accuracy: 0.9178\n", - "Epoch 62/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3068 - accuracy: 0.8659 - val_loss: 0.2609 - val_accuracy: 0.9269\n", - "Epoch 63/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3057 - accuracy: 0.8708 - val_loss: 0.2569 - val_accuracy: 0.9269\n", - "Epoch 64/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3022 - accuracy: 0.8787 - val_loss: 0.2704 - val_accuracy: 0.9132\n", - "Epoch 65/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.3012 - accuracy: 0.8767 - val_loss: 0.2571 - val_accuracy: 0.9224\n", - "Epoch 66/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2997 - accuracy: 0.8748 - val_loss: 0.2452 - val_accuracy: 0.9224\n", - "Epoch 67/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2977 - accuracy: 0.8796 - val_loss: 0.2413 - val_accuracy: 0.9087\n", - "Epoch 68/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2963 - accuracy: 0.8757 - val_loss: 0.2459 - val_accuracy: 0.9315\n", - "Epoch 69/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2969 - accuracy: 0.8728 - val_loss: 0.2496 - val_accuracy: 0.9224\n", - "Epoch 70/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2958 - accuracy: 0.8796 - val_loss: 0.2432 - val_accuracy: 0.9315\n", - "Epoch 71/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2924 - accuracy: 0.8826 - val_loss: 0.2363 - val_accuracy: 0.9315\n", - "Epoch 72/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2917 - accuracy: 0.8738 - val_loss: 0.2329 - val_accuracy: 0.9361\n", - "Epoch 73/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2916 - accuracy: 0.8787 - val_loss: 0.2372 - val_accuracy: 0.9315\n", - "Epoch 74/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2906 - accuracy: 0.8738 - val_loss: 0.2329 - val_accuracy: 0.9361\n", - "Epoch 75/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2871 - accuracy: 0.8787 - val_loss: 0.2273 - val_accuracy: 0.9178\n", - "Epoch 76/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2875 - accuracy: 0.8748 - val_loss: 0.2291 - val_accuracy: 0.9361\n", - "Epoch 77/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2865 - accuracy: 0.8806 - val_loss: 0.2399 - val_accuracy: 0.9224\n", - "Epoch 78/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2854 - accuracy: 0.8836 - val_loss: 0.2240 - val_accuracy: 0.9361\n", - "Epoch 79/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2821 - accuracy: 0.8826 - val_loss: 0.2317 - val_accuracy: 0.9269\n", - "Epoch 80/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2816 - accuracy: 0.8816 - val_loss: 0.2363 - val_accuracy: 0.9224\n", - "Epoch 81/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2818 - accuracy: 0.8845 - val_loss: 0.2268 - val_accuracy: 0.9224\n", - "Epoch 82/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2796 - accuracy: 0.8826 - val_loss: 0.2245 - val_accuracy: 0.9269\n", - "Epoch 83/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2792 - accuracy: 0.8855 - val_loss: 0.2217 - val_accuracy: 0.9361\n", - "Epoch 84/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2775 - accuracy: 0.8865 - val_loss: 0.2163 - val_accuracy: 0.9361\n", - "Epoch 85/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2789 - accuracy: 0.8796 - val_loss: 0.2169 - val_accuracy: 0.9361\n", - "Epoch 86/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2773 - accuracy: 0.8826 - val_loss: 0.2166 - val_accuracy: 0.9361\n", - "Epoch 87/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2754 - accuracy: 0.8845 - val_loss: 0.2183 - val_accuracy: 0.9361\n", - "Epoch 88/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2738 - accuracy: 0.8855 - val_loss: 0.2123 - val_accuracy: 0.9315\n", - "Epoch 89/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2738 - accuracy: 0.8836 - val_loss: 0.2106 - val_accuracy: 0.9361\n", - "Epoch 90/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2742 - accuracy: 0.8885 - val_loss: 0.2165 - val_accuracy: 0.9315\n", - "Epoch 91/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2706 - accuracy: 0.8826 - val_loss: 0.2089 - val_accuracy: 0.9269\n", - "Epoch 92/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2719 - accuracy: 0.8924 - val_loss: 0.2183 - val_accuracy: 0.9269\n", - "Epoch 93/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2701 - accuracy: 0.8885 - val_loss: 0.2074 - val_accuracy: 0.9315\n", - "Epoch 94/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2672 - accuracy: 0.8875 - val_loss: 0.2480 - val_accuracy: 0.9087\n", - "Epoch 95/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2716 - accuracy: 0.8865 - val_loss: 0.2108 - val_accuracy: 0.9361\n", - "Epoch 96/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2673 - accuracy: 0.8865 - val_loss: 0.2136 - val_accuracy: 0.9315\n", - "Epoch 97/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2681 - accuracy: 0.8875 - val_loss: 0.2105 - val_accuracy: 0.9315\n", - "Epoch 98/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2691 - accuracy: 0.8855 - val_loss: 0.2102 - val_accuracy: 0.9315\n", - "Epoch 99/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2659 - accuracy: 0.8875 - val_loss: 0.2103 - val_accuracy: 0.9315\n", - "Epoch 100/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2673 - accuracy: 0.8865 - val_loss: 0.2025 - val_accuracy: 0.9315\n", - "Epoch 101/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2648 - accuracy: 0.8845 - val_loss: 0.2089 - val_accuracy: 0.9315\n", - "Epoch 102/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2649 - accuracy: 0.8855 - val_loss: 0.2034 - val_accuracy: 0.9315\n", - "Epoch 103/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2630 - accuracy: 0.8855 - val_loss: 0.2155 - val_accuracy: 0.9224\n", - "Epoch 104/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2629 - accuracy: 0.8865 - val_loss: 0.2055 - val_accuracy: 0.9269\n", - "Epoch 105/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2649 - accuracy: 0.8836 - val_loss: 0.1999 - val_accuracy: 0.9269\n", - "Epoch 106/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2639 - accuracy: 0.8875 - val_loss: 0.2073 - val_accuracy: 0.9269\n", - "Epoch 107/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2619 - accuracy: 0.8855 - val_loss: 0.2007 - val_accuracy: 0.9315\n", - "Epoch 108/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2607 - accuracy: 0.8826 - val_loss: 0.1985 - val_accuracy: 0.9315\n", - "Epoch 109/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2608 - accuracy: 0.8894 - val_loss: 0.2164 - val_accuracy: 0.9269\n", - "Epoch 110/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2604 - accuracy: 0.8875 - val_loss: 0.2028 - val_accuracy: 0.9224\n", - "Epoch 111/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2630 - accuracy: 0.8885 - val_loss: 0.2008 - val_accuracy: 0.9315\n", - "Epoch 112/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2608 - accuracy: 0.8904 - val_loss: 0.2024 - val_accuracy: 0.9224\n", - "Epoch 113/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.2576 - accuracy: 0.8894 - val_loss: 0.1977 - val_accuracy: 0.9269\n", - "Epoch 114/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2599 - accuracy: 0.8924 - val_loss: 0.2090 - val_accuracy: 0.9178\n", - "Epoch 115/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2587 - accuracy: 0.8904 - val_loss: 0.2069 - val_accuracy: 0.9269\n", - "Epoch 116/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2584 - accuracy: 0.8924 - val_loss: 0.2109 - val_accuracy: 0.9224\n", - "Epoch 117/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2594 - accuracy: 0.8914 - val_loss: 0.1964 - val_accuracy: 0.9269\n", - "Epoch 118/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2572 - accuracy: 0.8885 - val_loss: 0.1954 - val_accuracy: 0.9269\n", - "Epoch 119/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2564 - accuracy: 0.8845 - val_loss: 0.1950 - val_accuracy: 0.9315\n", - "Epoch 120/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2545 - accuracy: 0.8875 - val_loss: 0.1966 - val_accuracy: 0.9224\n", - "Epoch 121/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2565 - accuracy: 0.8865 - val_loss: 0.2184 - val_accuracy: 0.9269\n", - "Epoch 122/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2577 - accuracy: 0.8865 - val_loss: 0.2005 - val_accuracy: 0.9224\n", - "Epoch 123/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2549 - accuracy: 0.8875 - val_loss: 0.1941 - val_accuracy: 0.9315\n", - "Epoch 124/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2573 - accuracy: 0.8855 - val_loss: 0.1946 - val_accuracy: 0.9224\n", - "Epoch 125/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2568 - accuracy: 0.8894 - val_loss: 0.1991 - val_accuracy: 0.9224\n", - "Epoch 126/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2538 - accuracy: 0.8904 - val_loss: 0.1949 - val_accuracy: 0.9224\n", - "Epoch 127/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2530 - accuracy: 0.8865 - val_loss: 0.1947 - val_accuracy: 0.9224\n", - "Epoch 128/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2549 - accuracy: 0.8885 - val_loss: 0.1933 - val_accuracy: 0.9269\n", - "Epoch 129/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.2544 - accuracy: 0.8914 - val_loss: 0.1972 - val_accuracy: 0.9178\n", - "Epoch 130/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.2534 - accuracy: 0.8885 - val_loss: 0.2089 - val_accuracy: 0.9224\n", - "Epoch 131/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2538 - accuracy: 0.8855 - val_loss: 0.1932 - val_accuracy: 0.9269\n", - "Epoch 132/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2531 - accuracy: 0.8855 - val_loss: 0.1948 - val_accuracy: 0.9178\n", - "Epoch 133/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2518 - accuracy: 0.8855 - val_loss: 0.1982 - val_accuracy: 0.9178\n", - "Epoch 134/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2529 - accuracy: 0.8836 - val_loss: 0.1972 - val_accuracy: 0.9178\n", - "Epoch 135/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2535 - accuracy: 0.8885 - val_loss: 0.1960 - val_accuracy: 0.9178\n", - "Epoch 136/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2507 - accuracy: 0.8875 - val_loss: 0.1925 - val_accuracy: 0.9224\n", - "Epoch 137/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2529 - accuracy: 0.8894 - val_loss: 0.1933 - val_accuracy: 0.9224\n", - "Epoch 138/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2526 - accuracy: 0.8855 - val_loss: 0.1963 - val_accuracy: 0.9178\n", - "Epoch 139/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2522 - accuracy: 0.8885 - val_loss: 0.2069 - val_accuracy: 0.9269\n", - "Epoch 140/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2508 - accuracy: 0.8885 - val_loss: 0.2062 - val_accuracy: 0.9224\n", - "Epoch 141/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2532 - accuracy: 0.8914 - val_loss: 0.1967 - val_accuracy: 0.9178\n", - "Epoch 142/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2504 - accuracy: 0.8875 - val_loss: 0.1930 - val_accuracy: 0.9178\n", - "Epoch 143/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2525 - accuracy: 0.8826 - val_loss: 0.2008 - val_accuracy: 0.9178\n", - "Epoch 144/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2522 - accuracy: 0.8865 - val_loss: 0.1934 - val_accuracy: 0.9178\n", - "Epoch 145/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2517 - accuracy: 0.8924 - val_loss: 0.1984 - val_accuracy: 0.9178\n", - "Epoch 146/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2509 - accuracy: 0.8924 - val_loss: 0.1980 - val_accuracy: 0.9178\n", - "Epoch 147/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2515 - accuracy: 0.8845 - val_loss: 0.1990 - val_accuracy: 0.9132\n", - "Epoch 148/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2506 - accuracy: 0.8855 - val_loss: 0.1926 - val_accuracy: 0.9178\n", - "Epoch 149/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2505 - accuracy: 0.8875 - val_loss: 0.2005 - val_accuracy: 0.9132\n", - "Epoch 150/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2495 - accuracy: 0.8924 - val_loss: 0.2105 - val_accuracy: 0.9178\n", - "Epoch 151/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2495 - accuracy: 0.8885 - val_loss: 0.1924 - val_accuracy: 0.9178\n", - "Epoch 152/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2492 - accuracy: 0.8855 - val_loss: 0.1907 - val_accuracy: 0.9269\n", - "Epoch 153/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2495 - accuracy: 0.8914 - val_loss: 0.1958 - val_accuracy: 0.9178\n", - "Epoch 154/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2493 - accuracy: 0.8855 - val_loss: 0.2010 - val_accuracy: 0.9132\n", - "Epoch 155/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2488 - accuracy: 0.8875 - val_loss: 0.1904 - val_accuracy: 0.9269\n", - "Epoch 156/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2493 - accuracy: 0.8875 - val_loss: 0.1931 - val_accuracy: 0.9178\n", - "Epoch 157/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2464 - accuracy: 0.8914 - val_loss: 0.2023 - val_accuracy: 0.9178\n", - "Epoch 158/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2489 - accuracy: 0.8865 - val_loss: 0.1931 - val_accuracy: 0.9178\n", - "Epoch 159/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2454 - accuracy: 0.8933 - val_loss: 0.2190 - val_accuracy: 0.9087\n", - "Epoch 160/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2487 - accuracy: 0.8933 - val_loss: 0.1959 - val_accuracy: 0.9178\n", - "Epoch 161/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2464 - accuracy: 0.8865 - val_loss: 0.1904 - val_accuracy: 0.9269\n", - "Epoch 162/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2497 - accuracy: 0.8875 - val_loss: 0.1903 - val_accuracy: 0.9269\n", - "Epoch 163/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2470 - accuracy: 0.8943 - val_loss: 0.1921 - val_accuracy: 0.9178\n", - "Epoch 164/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2482 - accuracy: 0.8845 - val_loss: 0.1912 - val_accuracy: 0.9224\n", - "Epoch 165/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2472 - accuracy: 0.8953 - val_loss: 0.1899 - val_accuracy: 0.9224\n", - "Epoch 166/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2492 - accuracy: 0.8894 - val_loss: 0.1910 - val_accuracy: 0.9224\n", - "Epoch 167/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2471 - accuracy: 0.8894 - val_loss: 0.1906 - val_accuracy: 0.9224\n", - "Epoch 168/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2451 - accuracy: 0.8875 - val_loss: 0.2076 - val_accuracy: 0.9269\n", - "Epoch 169/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2488 - accuracy: 0.8904 - val_loss: 0.1912 - val_accuracy: 0.9224\n", - "Epoch 170/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2457 - accuracy: 0.8875 - val_loss: 0.1915 - val_accuracy: 0.9178\n", - "Epoch 171/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2465 - accuracy: 0.8992 - val_loss: 0.1931 - val_accuracy: 0.9178\n", - "Epoch 172/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2517 - accuracy: 0.8855 - val_loss: 0.2003 - val_accuracy: 0.9087\n", - "Epoch 173/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2457 - accuracy: 0.8894 - val_loss: 0.1906 - val_accuracy: 0.9269\n", - "Epoch 174/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2463 - accuracy: 0.8904 - val_loss: 0.1896 - val_accuracy: 0.9224\n", - "Epoch 175/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2485 - accuracy: 0.8894 - val_loss: 0.1933 - val_accuracy: 0.9178\n", - "Epoch 176/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2468 - accuracy: 0.8943 - val_loss: 0.1900 - val_accuracy: 0.9224\n", - "Epoch 177/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2438 - accuracy: 0.8943 - val_loss: 0.2056 - val_accuracy: 0.9224\n", - "Epoch 178/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2470 - accuracy: 0.8904 - val_loss: 0.1911 - val_accuracy: 0.9224\n", - "Epoch 179/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2458 - accuracy: 0.8875 - val_loss: 0.1960 - val_accuracy: 0.9132\n", - "Epoch 180/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2452 - accuracy: 0.8885 - val_loss: 0.1936 - val_accuracy: 0.9178\n", - "Epoch 181/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2463 - accuracy: 0.8885 - val_loss: 0.1895 - val_accuracy: 0.9224\n", - "Epoch 182/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2427 - accuracy: 0.8963 - val_loss: 0.1908 - val_accuracy: 0.9224\n", - "Epoch 183/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2486 - accuracy: 0.8914 - val_loss: 0.1960 - val_accuracy: 0.9132\n", - "Epoch 184/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2439 - accuracy: 0.8924 - val_loss: 0.1889 - val_accuracy: 0.9224\n", - "Epoch 185/200\n", - "32/32 [==============================] - 0s 3ms/step - loss: 0.2471 - accuracy: 0.8904 - val_loss: 0.1893 - val_accuracy: 0.9224\n", - "Epoch 186/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2451 - accuracy: 0.8924 - val_loss: 0.1953 - val_accuracy: 0.9132\n", - "Epoch 187/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2465 - accuracy: 0.8943 - val_loss: 0.2014 - val_accuracy: 0.9132\n", - "Epoch 188/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2454 - accuracy: 0.8894 - val_loss: 0.1893 - val_accuracy: 0.9269\n", - "Epoch 189/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2459 - accuracy: 0.8904 - val_loss: 0.1932 - val_accuracy: 0.9178\n", - "Epoch 190/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2428 - accuracy: 0.8924 - val_loss: 0.1894 - val_accuracy: 0.9269\n", - "Epoch 191/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2450 - accuracy: 0.8904 - val_loss: 0.1926 - val_accuracy: 0.9178\n", - "Epoch 192/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2444 - accuracy: 0.8865 - val_loss: 0.1900 - val_accuracy: 0.9224\n", - "Epoch 193/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2448 - accuracy: 0.8894 - val_loss: 0.1962 - val_accuracy: 0.9087\n", - "Epoch 194/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2435 - accuracy: 0.8914 - val_loss: 0.1980 - val_accuracy: 0.9087\n", - "Epoch 195/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2446 - accuracy: 0.8914 - val_loss: 0.1927 - val_accuracy: 0.9178\n", - "Epoch 196/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2417 - accuracy: 0.8933 - val_loss: 0.1946 - val_accuracy: 0.9087\n", - "Epoch 197/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2434 - accuracy: 0.8865 - val_loss: 0.1899 - val_accuracy: 0.9224\n", - "Epoch 198/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2443 - accuracy: 0.8914 - val_loss: 0.1888 - val_accuracy: 0.9269\n", - "Epoch 199/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2453 - accuracy: 0.8992 - val_loss: 0.1894 - val_accuracy: 0.9224\n", - "Epoch 200/200\n", - "32/32 [==============================] - 0s 2ms/step - loss: 0.2427 - accuracy: 0.8894 - val_loss: 0.1980 - val_accuracy: 0.9087\n" - ] - } - ], - "source": [ - "from keras.models import Sequential\n", - "\n", - "from keras.layers import Dense\n", - "\n", - "model = Sequential([\n", - "Dense(32, activation='relu', input_shape=(10,)),\n", - "Dense(32, activation='relu'),\n", - "Dense(32, activation='relu'),\n", - "Dense(1, activation='sigmoid')\n", - "])\n", - "\n", - "model.compile(optimizer='sgd', loss='binary_crossentropy', metrics=['accuracy'])\n", - "model_hist = model.fit(X_training, Y_training, epochs=200, batch_size=32, validation_data=(X_valid, Y_valid))" - ] - }, { "cell_type": "markdown", "metadata": {}, @@ -807,135 +46,12 @@ "6. Testen der Genauigkeit der Vorhersagen" ] }, - { - "cell_type": "code", - "execution_count": 9, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "7/7 [==============================] - 0s 2ms/step - loss: 0.2772 - accuracy: 0.8858\n" - ] - }, - { - "data": { - "text/plain": [ - "0.8858447670936584" - ] - }, - "execution_count": 9, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "model.evaluate(X_test, Y_test)[1]" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Dritte Immobilien im Test-Array: Richtiges Ergebnis" - ] - }, - { - "cell_type": "code", - "execution_count": 10, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "0\n" - ] - } - ], - "source": [ - "print(Y_test[2])" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "Dritte Immobilien im Test-Array: Vorhersage" - ] - }, - { - "cell_type": "code", - "execution_count": 11, - "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "7/7 [==============================] - 0s 1ms/step\n" - ] - }, - { - "data": { - "text/plain": [ - "array([0.01198033], dtype=float32)" - ] - }, - "execution_count": 11, - "metadata": {}, - "output_type": "execute_result" - } - ], - "source": [ - "Y_test_vorhersage = model.predict(X_test)\n", - "\n", - "Y_test_vorhersage[2]" - ] - }, { "cell_type": "markdown", "metadata": {}, "source": [ "7. Visualisierung der Ergebnisse mit matplotlib" ] - }, - { - "cell_type": "code", - "execution_count": 12, - "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "<Figure size 1000x600 with 1 Axes>" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "from matplotlib import pyplot as plt\n", - "\n", - "plt.figure(figsize=(10,6))\n", - "\n", - "plt.plot(model_hist.history['accuracy'])\n", - "\n", - "plt.plot(model_hist.history['val_accuracy'])\n", - "\n", - "plt.title('Genauigkeit des Modells')\n", - "\n", - "plt.xlabel('Epochen')\n", - "\n", - "plt.ylabel('Genauigkeit')\n", - "\n", - "plt.legend(['Training', 'Validierung'])\n", - "\n", - "plt.show()" - ] } ], "metadata": {