ium_464903/dockerfiles/Biblioteka_DL_trenowanie.ipynb

1493 lines
142 KiB
Plaintext

{
"cells": [
{
"cell_type": "markdown",
"id": "12c11089-7d45-4e24-b797-83d3ae4841fa",
"metadata": {},
"source": [
"## Import bibliotek"
]
},
{
"cell_type": "code",
"execution_count": 1,
"id": "64cea583-a329-4df4-be36-18d94f15966d",
"metadata": {},
"outputs": [],
"source": [
"import tensorflow as tf\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.preprocessing import MinMaxScaler, OneHotEncoder\n",
"from keras.models import Sequential\n",
"from keras.layers import Dense, Dropout\n",
"from keras import regularizers\n",
"import numpy as np\n",
"import pandas as pd\n",
"import opendatasets as od\n",
"import chardet\n",
"import matplotlib.pyplot as plt"
]
},
{
"cell_type": "markdown",
"id": "f7a16131-7e52-43b8-96a6-694e3da8ccb1",
"metadata": {},
"source": [
"## Pobranie zbioru danych z kaggle"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "975e8b6a-4392-4fca-a4b8-8483c8782643",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Skipping, found downloaded files in \".\\lettuce-growth-days\" (use force=True to force download)\n"
]
}
],
"source": [
"od.download('https://www.kaggle.com/datasets/jjayfabor/lettuce-growth-days')\n",
"#{\"username\":\"jakubbg\",\"key\":\"e42b293c818e4ecd7b9365ee037af428\"}"
]
},
{
"cell_type": "markdown",
"id": "8ac9c0ef-250d-4d6c-93e2-fc533e665836",
"metadata": {},
"source": [
"## Załadowanie zbioru danych"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "45cd2f54-f973-4a02-b40a-369e45c51522",
"metadata": {},
"outputs": [],
"source": [
"with open('./lettuce-growth-days/lettuce_dataset_updated.csv', 'rb') as f:\n",
" result = chardet.detect(f.read())"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "57e8a928-a7a6-4e83-92b4-ffa19b04265e",
"metadata": {},
"outputs": [],
"source": [
"dataset = pd.read_csv('./lettuce-growth-days/lettuce_dataset_updated.csv', encoding=result['encoding'])"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "add14d7c-dc88-42dd-9113-b9dc0e9b5c83",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"3169\n",
" Plant_ID Date Temperature (°C) Humidity (%) TDS Value (ppm) \\\n",
"0 1 8/3/2023 33.4 53 582 \n",
"1 1 8/4/2023 33.5 53 451 \n",
"2 1 8/5/2023 33.4 59 678 \n",
"3 1 8/6/2023 33.4 68 420 \n",
"4 1 8/7/2023 33.4 74 637 \n",
"\n",
" pH Level Growth Days Temperature (F) Humidity \n",
"0 6.4 1 92.12 0.53 \n",
"1 6.1 2 92.30 0.53 \n",
"2 6.4 3 92.12 0.59 \n",
"3 6.4 4 92.12 0.68 \n",
"4 6.5 5 92.12 0.74 \n"
]
}
],
"source": [
"print(len(dataset))\n",
"print(dataset[:5])"
]
},
{
"cell_type": "markdown",
"id": "e9e15b0b-4f5d-4135-a4aa-e7160cfd8292",
"metadata": {},
"source": [
"## Wyciąganie ze zbioru wybranych kolumn"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "e9a90813-275d-466a-936f-0d427d442319",
"metadata": {},
"outputs": [],
"source": [
"ph_level = dataset['pH Level'].values.tolist()\n",
"temp_F = dataset['Temperature (F)'].values.tolist()\n",
"humid = dataset['Humidity'].values.tolist()\n",
"days = dataset['Growth Days'].values.tolist()\n",
"plant_id = dataset['Plant_ID'].values.tolist()"
]
},
{
"cell_type": "markdown",
"id": "636b9e9c-fa7d-4408-932c-c7a1648b251e",
"metadata": {},
"source": [
"## Przetwarzanie danych do postaci zbioru X i Y"
]
},
{
"cell_type": "markdown",
"id": "b9135091-4559-4ed7-b2ba-36dc82529019",
"metadata": {},
"source": [
"Każda próbka składać się będzie ze średniej temperatury (F), średniej wilgotności oraz średniego ph gleby dla danej rośliny oraz z przypisanej jej klasy będącej ilością dni, jakie były wymagane do całkowitego wyrośnięcia rośliny."
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "c91849f4-2e8e-4eba-9b65-824f045e7ddb",
"metadata": {},
"outputs": [],
"source": [
"X = []\n",
"Y = []\n",
"\n",
"id = plant_id[0]\n",
"temp_sum = 0\n",
"humid_sum = 0\n",
"ph_level_sum = 0\n",
"day = 1\n",
"\n",
"for i in range(0, len(plant_id)):\n",
" if plant_id[i] == id:\n",
" temp_sum += temp_F[i]\n",
" humid_sum += humid[i]\n",
" ph_level_sum += ph_level[i]\n",
" day = days[i]\n",
" else:\n",
" temp = []\n",
" temp.append(temp_sum/day)\n",
" temp.append(humid_sum/day)\n",
" temp.append(ph_level_sum/day)\n",
" X.append(temp)\n",
" Y.append(day)\n",
" temp_sum = 0\n",
" humid_sum = 0\n",
" ph_level_sum = 0\n",
" day = 1\n",
" id = plant_id[i]"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "d8e7682d-3469-4d68-98e0-f34b1d57ae27",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[[87.13199999999998, 0.6395555555555558, 6.382222222222223], [85.08488888888886, 0.6295555555555553, 6.311111111111113], [85.13148936170211, 0.6759574468085107, 6.27659574468085], [85.24333333333331, 0.6206250000000001, 6.293749999999999], [85.08488888888886, 0.6357777777777776, 6.264444444444444], [85.08488888888886, 0.6295555555555555, 6.297777777777776], [85.10851063829786, 0.625531914893617, 6.285106382978723], [85.08488888888886, 0.6457777777777778, 6.266666666666666], [85.15565217391303, 0.6530434782608696, 6.2195652173913025], [85.08488888888886, 0.6435555555555555, 6.262222222222222]]\n"
]
}
],
"source": [
"print(X[:10])"
]
},
{
"cell_type": "markdown",
"id": "a6e7e9a2-7521-4aea-9e37-36256800e064",
"metadata": {},
"source": [
"## Normalizacja danych "
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "ca14eeb3-6616-405b-98a9-373e3ee8e07d",
"metadata": {},
"outputs": [],
"source": [
"scaler = MinMaxScaler()\n",
"X = scaler.fit_transform(X)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "adfcfa06-d323-4c76-aaaa-cd2b98aeb18a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"(41, 3) (28, 3) (41, 4) (28, 4)\n"
]
},
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\obses\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\sklearn\\preprocessing\\_encoders.py:808: FutureWarning: `sparse` was renamed to `sparse_output` in version 1.2 and will be removed in 1.4. `sparse_output` is ignored unless you leave `sparse` to its default value.\n",
" warnings.warn(\n"
]
}
],
"source": [
"X = np.array(X)\n",
"Y = np.array(Y)\n",
"\n",
"encoder = OneHotEncoder(sparse=False)\n",
"y_onehot = encoder.fit_transform(Y.reshape(-1,1))\n",
"\n",
"X_train, X_test, y_train, y_test = train_test_split(X, y_onehot, test_size=0.4, random_state=42)\n",
"print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)"
]
},
{
"cell_type": "markdown",
"id": "83689322-9d41-46db-9fbd-6f3040b3249c",
"metadata": {},
"source": [
"## Budowanie modelu sieci neuronowej"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "df2e1035-ae75-4fbf-9216-2738ed4cadd0",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"C:\\Users\\obses\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\keras\\src\\layers\\core\\dense.py:86: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
" super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n"
]
}
],
"source": [
"model = Sequential([\n",
" Dense(8, activation='relu', input_dim=3, kernel_regularizer=regularizers.l2(0.04)),\n",
" Dropout(0.5),\n",
" Dense(8, activation='relu', kernel_regularizer=regularizers.l2(0.04)),\n",
" Dropout(0.5),\n",
" Dense(4, activation='softmax', kernel_regularizer=regularizers.l2(0.04)),\n",
"])"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "0ff7bb18-e65f-4891-9ace-d3ab02b2325d",
"metadata": {},
"outputs": [],
"source": [
"model.compile(optimizer='sgd',\n",
" loss='categorical_crossentropy',\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "markdown",
"id": "2ac98bdc-1851-439d-8306-4fe651a70069",
"metadata": {},
"source": [
"## Trenowanie modelu"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "9b89b3c2-24de-4127-9aba-3e8402f09b5f",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/500\n",
"2/2 - 1s - 265ms/step - accuracy: 0.3902 - loss: 2.1238 - val_accuracy: 0.1429 - val_loss: 2.1243\n",
"Epoch 2/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.2439 - loss: 2.1641 - val_accuracy: 0.2500 - val_loss: 2.1104\n",
"Epoch 3/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.3902 - loss: 2.1102 - val_accuracy: 0.5714 - val_loss: 2.0970\n",
"Epoch 4/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.4146 - loss: 2.0846 - val_accuracy: 0.7857 - val_loss: 2.0847\n",
"Epoch 5/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.5854 - loss: 2.1145 - val_accuracy: 0.8571 - val_loss: 2.0710\n",
"Epoch 6/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.6585 - loss: 2.0605 - val_accuracy: 0.8571 - val_loss: 2.0591\n",
"Epoch 7/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.6829 - loss: 2.0652 - val_accuracy: 0.8571 - val_loss: 2.0459\n",
"Epoch 8/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.6585 - loss: 2.0399 - val_accuracy: 0.8571 - val_loss: 2.0329\n",
"Epoch 9/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8049 - loss: 2.0242 - val_accuracy: 0.8571 - val_loss: 2.0210\n",
"Epoch 10/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.6829 - loss: 2.0091 - val_accuracy: 0.8571 - val_loss: 2.0091\n",
"Epoch 11/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8049 - loss: 1.9942 - val_accuracy: 0.8571 - val_loss: 1.9976\n",
"Epoch 12/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7561 - loss: 1.9887 - val_accuracy: 0.8571 - val_loss: 1.9867\n",
"Epoch 13/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.7805 - loss: 1.9710 - val_accuracy: 0.8571 - val_loss: 1.9748\n",
"Epoch 14/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7805 - loss: 1.9850 - val_accuracy: 0.8571 - val_loss: 1.9635\n",
"Epoch 15/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7805 - loss: 1.9553 - val_accuracy: 0.8571 - val_loss: 1.9527\n",
"Epoch 16/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.7805 - loss: 1.9349 - val_accuracy: 0.8571 - val_loss: 1.9419\n",
"Epoch 17/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7561 - loss: 1.9401 - val_accuracy: 0.8571 - val_loss: 1.9307\n",
"Epoch 18/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7805 - loss: 1.9213 - val_accuracy: 0.8571 - val_loss: 1.9209\n",
"Epoch 19/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8293 - loss: 1.9173 - val_accuracy: 0.8571 - val_loss: 1.9100\n",
"Epoch 20/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8293 - loss: 1.8995 - val_accuracy: 0.8571 - val_loss: 1.8995\n",
"Epoch 21/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8780 - loss: 1.8758 - val_accuracy: 0.8571 - val_loss: 1.8900\n",
"Epoch 22/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8293 - loss: 1.8806 - val_accuracy: 0.8571 - val_loss: 1.8796\n",
"Epoch 23/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8293 - loss: 1.8815 - val_accuracy: 0.8571 - val_loss: 1.8693\n",
"Epoch 24/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8293 - loss: 1.8560 - val_accuracy: 0.8571 - val_loss: 1.8589\n",
"Epoch 25/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.7805 - loss: 1.8495 - val_accuracy: 0.8571 - val_loss: 1.8491\n",
"Epoch 26/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.8245 - val_accuracy: 0.8571 - val_loss: 1.8394\n",
"Epoch 27/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8049 - loss: 1.8372 - val_accuracy: 0.8571 - val_loss: 1.8294\n",
"Epoch 28/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.7805 - loss: 1.8236 - val_accuracy: 0.8571 - val_loss: 1.8191\n",
"Epoch 29/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8293 - loss: 1.8232 - val_accuracy: 0.8571 - val_loss: 1.8090\n",
"Epoch 30/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.7873 - val_accuracy: 0.8571 - val_loss: 1.8008\n",
"Epoch 31/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8293 - loss: 1.7862 - val_accuracy: 0.8571 - val_loss: 1.7918\n",
"Epoch 32/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.8029 - val_accuracy: 0.8571 - val_loss: 1.7828\n",
"Epoch 33/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.7658 - val_accuracy: 0.8571 - val_loss: 1.7734\n",
"Epoch 34/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.7646 - val_accuracy: 0.8571 - val_loss: 1.7637\n",
"Epoch 35/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.7255 - val_accuracy: 0.8571 - val_loss: 1.7539\n",
"Epoch 36/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.7385 - val_accuracy: 0.8571 - val_loss: 1.7438\n",
"Epoch 37/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8293 - loss: 1.7172 - val_accuracy: 0.8571 - val_loss: 1.7340\n",
"Epoch 38/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.7158 - val_accuracy: 0.8571 - val_loss: 1.7246\n",
"Epoch 39/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.7122 - val_accuracy: 0.8571 - val_loss: 1.7162\n",
"Epoch 40/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.7141 - val_accuracy: 0.8571 - val_loss: 1.7086\n",
"Epoch 41/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6996 - val_accuracy: 0.8571 - val_loss: 1.6996\n",
"Epoch 42/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.6795 - val_accuracy: 0.8571 - val_loss: 1.6900\n",
"Epoch 43/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.6662 - val_accuracy: 0.8571 - val_loss: 1.6833\n",
"Epoch 44/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.6729 - val_accuracy: 0.8571 - val_loss: 1.6753\n",
"Epoch 45/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6760 - val_accuracy: 0.8571 - val_loss: 1.6665\n",
"Epoch 46/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6566 - val_accuracy: 0.8571 - val_loss: 1.6581\n",
"Epoch 47/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6475 - val_accuracy: 0.8571 - val_loss: 1.6501\n",
"Epoch 48/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6537 - val_accuracy: 0.8571 - val_loss: 1.6430\n",
"Epoch 49/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.6250 - val_accuracy: 0.8571 - val_loss: 1.6356\n",
"Epoch 50/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6135 - val_accuracy: 0.8571 - val_loss: 1.6266\n",
"Epoch 51/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.6359 - val_accuracy: 0.8571 - val_loss: 1.6192\n",
"Epoch 52/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.5970 - val_accuracy: 0.8571 - val_loss: 1.6120\n",
"Epoch 53/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.6041 - val_accuracy: 0.8571 - val_loss: 1.6046\n",
"Epoch 54/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.5999 - val_accuracy: 0.8571 - val_loss: 1.5963\n",
"Epoch 55/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 1.5691 - val_accuracy: 0.8571 - val_loss: 1.5882\n",
"Epoch 56/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.5940 - val_accuracy: 0.8571 - val_loss: 1.5820\n",
"Epoch 57/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.5604 - val_accuracy: 0.8571 - val_loss: 1.5755\n",
"Epoch 58/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5783 - val_accuracy: 0.8571 - val_loss: 1.5676\n",
"Epoch 59/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.5775 - val_accuracy: 0.8571 - val_loss: 1.5591\n",
"Epoch 60/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5195 - val_accuracy: 0.8571 - val_loss: 1.5509\n",
"Epoch 61/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5291 - val_accuracy: 0.8571 - val_loss: 1.5442\n",
"Epoch 62/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5216 - val_accuracy: 0.8571 - val_loss: 1.5362\n",
"Epoch 63/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.5330 - val_accuracy: 0.8571 - val_loss: 1.5293\n",
"Epoch 64/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5344 - val_accuracy: 0.8571 - val_loss: 1.5228\n",
"Epoch 65/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5161 - val_accuracy: 0.8571 - val_loss: 1.5158\n",
"Epoch 66/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4849 - val_accuracy: 0.8571 - val_loss: 1.5081\n",
"Epoch 67/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.5029 - val_accuracy: 0.8571 - val_loss: 1.5009\n",
"Epoch 68/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.4785 - val_accuracy: 0.8571 - val_loss: 1.4936\n",
"Epoch 69/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.5226 - val_accuracy: 0.8571 - val_loss: 1.4871\n",
"Epoch 70/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.4863 - val_accuracy: 0.8571 - val_loss: 1.4801\n",
"Epoch 71/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.4651 - val_accuracy: 0.8571 - val_loss: 1.4741\n",
"Epoch 72/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.4743 - val_accuracy: 0.8571 - val_loss: 1.4673\n",
"Epoch 73/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4487 - val_accuracy: 0.8571 - val_loss: 1.4599\n",
"Epoch 74/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4422 - val_accuracy: 0.8571 - val_loss: 1.4536\n",
"Epoch 75/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.4413 - val_accuracy: 0.8571 - val_loss: 1.4475\n",
"Epoch 76/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4394 - val_accuracy: 0.8571 - val_loss: 1.4413\n",
"Epoch 77/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.4215 - val_accuracy: 0.8571 - val_loss: 1.4355\n",
"Epoch 78/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.4270 - val_accuracy: 0.8571 - val_loss: 1.4285\n",
"Epoch 79/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.4395 - val_accuracy: 0.8571 - val_loss: 1.4219\n",
"Epoch 80/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.4311 - val_accuracy: 0.8571 - val_loss: 1.4154\n",
"Epoch 81/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4002 - val_accuracy: 0.8571 - val_loss: 1.4086\n",
"Epoch 82/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.4201 - val_accuracy: 0.8571 - val_loss: 1.4030\n",
"Epoch 83/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3839 - val_accuracy: 0.8571 - val_loss: 1.3973\n",
"Epoch 84/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.4089 - val_accuracy: 0.8571 - val_loss: 1.3913\n",
"Epoch 85/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3913 - val_accuracy: 0.8571 - val_loss: 1.3845\n",
"Epoch 86/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.3826 - val_accuracy: 0.8571 - val_loss: 1.3788\n",
"Epoch 87/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3724 - val_accuracy: 0.8571 - val_loss: 1.3733\n",
"Epoch 88/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3838 - val_accuracy: 0.8571 - val_loss: 1.3679\n",
"Epoch 89/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3502 - val_accuracy: 0.8571 - val_loss: 1.3624\n",
"Epoch 90/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.3619 - val_accuracy: 0.8571 - val_loss: 1.3573\n",
"Epoch 91/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3734 - val_accuracy: 0.8571 - val_loss: 1.3538\n",
"Epoch 92/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3499 - val_accuracy: 0.8571 - val_loss: 1.3479\n",
"Epoch 93/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3392 - val_accuracy: 0.8571 - val_loss: 1.3426\n",
"Epoch 94/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3334 - val_accuracy: 0.8571 - val_loss: 1.3370\n",
"Epoch 95/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3347 - val_accuracy: 0.8571 - val_loss: 1.3315\n",
"Epoch 96/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.3316 - val_accuracy: 0.8571 - val_loss: 1.3265\n",
"Epoch 97/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3638 - val_accuracy: 0.8571 - val_loss: 1.3228\n",
"Epoch 98/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3231 - val_accuracy: 0.8571 - val_loss: 1.3187\n",
"Epoch 99/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3332 - val_accuracy: 0.8571 - val_loss: 1.3138\n",
"Epoch 100/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.3142 - val_accuracy: 0.8571 - val_loss: 1.3088\n",
"Epoch 101/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3083 - val_accuracy: 0.8571 - val_loss: 1.3047\n",
"Epoch 102/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2839 - val_accuracy: 0.8571 - val_loss: 1.2998\n",
"Epoch 103/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.2872 - val_accuracy: 0.8571 - val_loss: 1.2956\n",
"Epoch 104/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.3188 - val_accuracy: 0.8571 - val_loss: 1.2922\n",
"Epoch 105/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.2763 - val_accuracy: 0.8571 - val_loss: 1.2873\n",
"Epoch 106/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.2957 - val_accuracy: 0.8571 - val_loss: 1.2831\n",
"Epoch 107/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 1.2953 - val_accuracy: 0.8571 - val_loss: 1.2787\n",
"Epoch 108/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2710 - val_accuracy: 0.8571 - val_loss: 1.2737\n",
"Epoch 109/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2953 - val_accuracy: 0.8571 - val_loss: 1.2693\n",
"Epoch 110/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.2716 - val_accuracy: 0.8571 - val_loss: 1.2660\n",
"Epoch 111/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2748 - val_accuracy: 0.8571 - val_loss: 1.2631\n",
"Epoch 112/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.2488 - val_accuracy: 0.8571 - val_loss: 1.2585\n",
"Epoch 113/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2603 - val_accuracy: 0.8571 - val_loss: 1.2538\n",
"Epoch 114/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2413 - val_accuracy: 0.8571 - val_loss: 1.2496\n",
"Epoch 115/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2690 - val_accuracy: 0.8571 - val_loss: 1.2452\n",
"Epoch 116/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2510 - val_accuracy: 0.8571 - val_loss: 1.2411\n",
"Epoch 117/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2378 - val_accuracy: 0.8571 - val_loss: 1.2377\n",
"Epoch 118/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.2281 - val_accuracy: 0.8571 - val_loss: 1.2339\n",
"Epoch 119/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2639 - val_accuracy: 0.8571 - val_loss: 1.2304\n",
"Epoch 120/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2622 - val_accuracy: 0.8571 - val_loss: 1.2262\n",
"Epoch 121/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.2354 - val_accuracy: 0.8571 - val_loss: 1.2230\n",
"Epoch 122/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.2722 - val_accuracy: 0.8571 - val_loss: 1.2196\n",
"Epoch 123/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.2421 - val_accuracy: 0.8571 - val_loss: 1.2153\n",
"Epoch 124/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2292 - val_accuracy: 0.8571 - val_loss: 1.2126\n",
"Epoch 125/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.2138 - val_accuracy: 0.8571 - val_loss: 1.2092\n",
"Epoch 126/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1895 - val_accuracy: 0.8571 - val_loss: 1.2055\n",
"Epoch 127/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.2061 - val_accuracy: 0.8571 - val_loss: 1.2014\n",
"Epoch 128/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1968 - val_accuracy: 0.8571 - val_loss: 1.1979\n",
"Epoch 129/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.2231 - val_accuracy: 0.8571 - val_loss: 1.1943\n",
"Epoch 130/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.2190 - val_accuracy: 0.8571 - val_loss: 1.1910\n",
"Epoch 131/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.2191 - val_accuracy: 0.8571 - val_loss: 1.1880\n",
"Epoch 132/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.2025 - val_accuracy: 0.8571 - val_loss: 1.1843\n",
"Epoch 133/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.1829 - val_accuracy: 0.8571 - val_loss: 1.1806\n",
"Epoch 134/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 1.1836 - val_accuracy: 0.8571 - val_loss: 1.1770\n",
"Epoch 135/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.1937 - val_accuracy: 0.8571 - val_loss: 1.1741\n",
"Epoch 136/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1785 - val_accuracy: 0.8571 - val_loss: 1.1710\n",
"Epoch 137/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1702 - val_accuracy: 0.8571 - val_loss: 1.1680\n",
"Epoch 138/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.1911 - val_accuracy: 0.8571 - val_loss: 1.1646\n",
"Epoch 139/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.1821 - val_accuracy: 0.8571 - val_loss: 1.1611\n",
"Epoch 140/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1455 - val_accuracy: 0.8571 - val_loss: 1.1576\n",
"Epoch 141/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1711 - val_accuracy: 0.8571 - val_loss: 1.1544\n",
"Epoch 142/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.1591 - val_accuracy: 0.8571 - val_loss: 1.1509\n",
"Epoch 143/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.1675 - val_accuracy: 0.8571 - val_loss: 1.1474\n",
"Epoch 144/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1860 - val_accuracy: 0.8571 - val_loss: 1.1444\n",
"Epoch 145/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.1656 - val_accuracy: 0.8571 - val_loss: 1.1419\n",
"Epoch 146/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1621 - val_accuracy: 0.8571 - val_loss: 1.1386\n",
"Epoch 147/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1650 - val_accuracy: 0.8571 - val_loss: 1.1365\n",
"Epoch 148/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.1870 - val_accuracy: 0.8571 - val_loss: 1.1336\n",
"Epoch 149/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1472 - val_accuracy: 0.8571 - val_loss: 1.1304\n",
"Epoch 150/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.1646 - val_accuracy: 0.8571 - val_loss: 1.1277\n",
"Epoch 151/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1335 - val_accuracy: 0.8571 - val_loss: 1.1259\n",
"Epoch 152/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1459 - val_accuracy: 0.8571 - val_loss: 1.1237\n",
"Epoch 153/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1264 - val_accuracy: 0.8571 - val_loss: 1.1209\n",
"Epoch 154/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1382 - val_accuracy: 0.8571 - val_loss: 1.1182\n",
"Epoch 155/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.1280 - val_accuracy: 0.8571 - val_loss: 1.1153\n",
"Epoch 156/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1313 - val_accuracy: 0.8571 - val_loss: 1.1123\n",
"Epoch 157/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1102 - val_accuracy: 0.8571 - val_loss: 1.1094\n",
"Epoch 158/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.1071 - val_accuracy: 0.8571 - val_loss: 1.1069\n",
"Epoch 159/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.0845 - val_accuracy: 0.8571 - val_loss: 1.1038\n",
"Epoch 160/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.1140 - val_accuracy: 0.8571 - val_loss: 1.1009\n",
"Epoch 161/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1390 - val_accuracy: 0.8571 - val_loss: 1.0985\n",
"Epoch 162/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0967 - val_accuracy: 0.8571 - val_loss: 1.0959\n",
"Epoch 163/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0894 - val_accuracy: 0.8571 - val_loss: 1.0935\n",
"Epoch 164/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.0937 - val_accuracy: 0.8571 - val_loss: 1.0908\n",
"Epoch 165/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1170 - val_accuracy: 0.8571 - val_loss: 1.0882\n",
"Epoch 166/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.1016 - val_accuracy: 0.8571 - val_loss: 1.0855\n",
"Epoch 167/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1365 - val_accuracy: 0.8571 - val_loss: 1.0842\n",
"Epoch 168/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.1163 - val_accuracy: 0.8571 - val_loss: 1.0821\n",
"Epoch 169/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0915 - val_accuracy: 0.8571 - val_loss: 1.0794\n",
"Epoch 170/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0816 - val_accuracy: 0.8571 - val_loss: 1.0777\n",
"Epoch 171/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1195 - val_accuracy: 0.8571 - val_loss: 1.0762\n",
"Epoch 172/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0741 - val_accuracy: 0.8571 - val_loss: 1.0741\n",
"Epoch 173/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.1333 - val_accuracy: 0.8571 - val_loss: 1.0718\n",
"Epoch 174/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0593 - val_accuracy: 0.8571 - val_loss: 1.0693\n",
"Epoch 175/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0957 - val_accuracy: 0.8571 - val_loss: 1.0670\n",
"Epoch 176/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0722 - val_accuracy: 0.8571 - val_loss: 1.0650\n",
"Epoch 177/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0786 - val_accuracy: 0.8571 - val_loss: 1.0626\n",
"Epoch 178/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0939 - val_accuracy: 0.8571 - val_loss: 1.0609\n",
"Epoch 179/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0600 - val_accuracy: 0.8571 - val_loss: 1.0586\n",
"Epoch 180/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.0768 - val_accuracy: 0.8571 - val_loss: 1.0561\n",
"Epoch 181/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0888 - val_accuracy: 0.8571 - val_loss: 1.0545\n",
"Epoch 182/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0809 - val_accuracy: 0.8571 - val_loss: 1.0526\n",
"Epoch 183/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0803 - val_accuracy: 0.8571 - val_loss: 1.0508\n",
"Epoch 184/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0355 - val_accuracy: 0.8571 - val_loss: 1.0486\n",
"Epoch 185/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0581 - val_accuracy: 0.8571 - val_loss: 1.0463\n",
"Epoch 186/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0827 - val_accuracy: 0.8571 - val_loss: 1.0447\n",
"Epoch 187/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0457 - val_accuracy: 0.8571 - val_loss: 1.0424\n",
"Epoch 188/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.0647 - val_accuracy: 0.8571 - val_loss: 1.0413\n",
"Epoch 189/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0579 - val_accuracy: 0.8571 - val_loss: 1.0397\n",
"Epoch 190/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0724 - val_accuracy: 0.8571 - val_loss: 1.0374\n",
"Epoch 191/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.0497 - val_accuracy: 0.8571 - val_loss: 1.0353\n",
"Epoch 192/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.0767 - val_accuracy: 0.8571 - val_loss: 1.0335\n",
"Epoch 193/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.0515 - val_accuracy: 0.8571 - val_loss: 1.0314\n",
"Epoch 194/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.0273 - val_accuracy: 0.8571 - val_loss: 1.0289\n",
"Epoch 195/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0470 - val_accuracy: 0.8571 - val_loss: 1.0268\n",
"Epoch 196/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0527 - val_accuracy: 0.8571 - val_loss: 1.0246\n",
"Epoch 197/500\n",
"2/2 - 0s - 18ms/step - accuracy: 0.8537 - loss: 1.0460 - val_accuracy: 0.8571 - val_loss: 1.0228\n",
"Epoch 198/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0396 - val_accuracy: 0.8571 - val_loss: 1.0212\n",
"Epoch 199/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0383 - val_accuracy: 0.8571 - val_loss: 1.0199\n",
"Epoch 200/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.0257 - val_accuracy: 0.8571 - val_loss: 1.0177\n",
"Epoch 201/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.0097 - val_accuracy: 0.8571 - val_loss: 1.0159\n",
"Epoch 202/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9978 - val_accuracy: 0.8571 - val_loss: 1.0139\n",
"Epoch 203/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0471 - val_accuracy: 0.8571 - val_loss: 1.0119\n",
"Epoch 204/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0409 - val_accuracy: 0.8571 - val_loss: 1.0102\n",
"Epoch 205/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0357 - val_accuracy: 0.8571 - val_loss: 1.0086\n",
"Epoch 206/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0442 - val_accuracy: 0.8571 - val_loss: 1.0075\n",
"Epoch 207/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0219 - val_accuracy: 0.8571 - val_loss: 1.0058\n",
"Epoch 208/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0257 - val_accuracy: 0.8571 - val_loss: 1.0042\n",
"Epoch 209/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9953 - val_accuracy: 0.8571 - val_loss: 1.0021\n",
"Epoch 210/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 1.0650 - val_accuracy: 0.8571 - val_loss: 1.0003\n",
"Epoch 211/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9909 - val_accuracy: 0.8571 - val_loss: 0.9984\n",
"Epoch 212/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9847 - val_accuracy: 0.8571 - val_loss: 0.9963\n",
"Epoch 213/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9929 - val_accuracy: 0.8571 - val_loss: 0.9942\n",
"Epoch 214/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9696 - val_accuracy: 0.8571 - val_loss: 0.9922\n",
"Epoch 215/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0115 - val_accuracy: 0.8571 - val_loss: 0.9909\n",
"Epoch 216/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0426 - val_accuracy: 0.8571 - val_loss: 0.9896\n",
"Epoch 217/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 1.0355 - val_accuracy: 0.8571 - val_loss: 0.9878\n",
"Epoch 218/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0111 - val_accuracy: 0.8571 - val_loss: 0.9862\n",
"Epoch 219/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0047 - val_accuracy: 0.8571 - val_loss: 0.9850\n",
"Epoch 220/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9901 - val_accuracy: 0.8571 - val_loss: 0.9831\n",
"Epoch 221/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.9875 - val_accuracy: 0.8571 - val_loss: 0.9815\n",
"Epoch 222/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9783 - val_accuracy: 0.8571 - val_loss: 0.9796\n",
"Epoch 223/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 1.0097 - val_accuracy: 0.8571 - val_loss: 0.9780\n",
"Epoch 224/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9895 - val_accuracy: 0.8571 - val_loss: 0.9764\n",
"Epoch 225/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 1.0003 - val_accuracy: 0.8571 - val_loss: 0.9752\n",
"Epoch 226/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 1.0074 - val_accuracy: 0.8571 - val_loss: 0.9736\n",
"Epoch 227/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.9722 - val_accuracy: 0.8571 - val_loss: 0.9722\n",
"Epoch 228/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9806 - val_accuracy: 0.8571 - val_loss: 0.9703\n",
"Epoch 229/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9936 - val_accuracy: 0.8571 - val_loss: 0.9686\n",
"Epoch 230/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9889 - val_accuracy: 0.8571 - val_loss: 0.9668\n",
"Epoch 231/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9880 - val_accuracy: 0.8571 - val_loss: 0.9654\n",
"Epoch 232/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9783 - val_accuracy: 0.8571 - val_loss: 0.9638\n",
"Epoch 233/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9825 - val_accuracy: 0.8571 - val_loss: 0.9623\n",
"Epoch 234/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9876 - val_accuracy: 0.8571 - val_loss: 0.9615\n",
"Epoch 235/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.0037 - val_accuracy: 0.8571 - val_loss: 0.9605\n",
"Epoch 236/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9587 - val_accuracy: 0.8571 - val_loss: 0.9587\n",
"Epoch 237/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9992 - val_accuracy: 0.8571 - val_loss: 0.9576\n",
"Epoch 238/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.9783 - val_accuracy: 0.8571 - val_loss: 0.9562\n",
"Epoch 239/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 1.0140 - val_accuracy: 0.8571 - val_loss: 0.9549\n",
"Epoch 240/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9395 - val_accuracy: 0.8571 - val_loss: 0.9534\n",
"Epoch 241/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9933 - val_accuracy: 0.8571 - val_loss: 0.9526\n",
"Epoch 242/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9540 - val_accuracy: 0.8571 - val_loss: 0.9511\n",
"Epoch 243/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9409 - val_accuracy: 0.8571 - val_loss: 0.9498\n",
"Epoch 244/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9850 - val_accuracy: 0.8571 - val_loss: 0.9487\n",
"Epoch 245/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9520 - val_accuracy: 0.8571 - val_loss: 0.9474\n",
"Epoch 246/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9655 - val_accuracy: 0.8571 - val_loss: 0.9462\n",
"Epoch 247/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9951 - val_accuracy: 0.8571 - val_loss: 0.9448\n",
"Epoch 248/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9752 - val_accuracy: 0.8571 - val_loss: 0.9434\n",
"Epoch 249/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9690 - val_accuracy: 0.8571 - val_loss: 0.9419\n",
"Epoch 250/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9625 - val_accuracy: 0.8571 - val_loss: 0.9407\n",
"Epoch 251/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9777 - val_accuracy: 0.8571 - val_loss: 0.9396\n",
"Epoch 252/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9895 - val_accuracy: 0.8571 - val_loss: 0.9383\n",
"Epoch 253/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9370 - val_accuracy: 0.8571 - val_loss: 0.9368\n",
"Epoch 254/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.9518 - val_accuracy: 0.8571 - val_loss: 0.9352\n",
"Epoch 255/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.9588 - val_accuracy: 0.8571 - val_loss: 0.9337\n",
"Epoch 256/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9775 - val_accuracy: 0.8571 - val_loss: 0.9326\n",
"Epoch 257/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.9639 - val_accuracy: 0.8571 - val_loss: 0.9314\n",
"Epoch 258/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9272 - val_accuracy: 0.8571 - val_loss: 0.9299\n",
"Epoch 259/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9601 - val_accuracy: 0.8571 - val_loss: 0.9284\n",
"Epoch 260/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9253 - val_accuracy: 0.8571 - val_loss: 0.9271\n",
"Epoch 261/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9673 - val_accuracy: 0.8571 - val_loss: 0.9258\n",
"Epoch 262/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9492 - val_accuracy: 0.8571 - val_loss: 0.9246\n",
"Epoch 263/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9503 - val_accuracy: 0.8571 - val_loss: 0.9232\n",
"Epoch 264/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.9502 - val_accuracy: 0.8571 - val_loss: 0.9223\n",
"Epoch 265/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8984 - val_accuracy: 0.8571 - val_loss: 0.9208\n",
"Epoch 266/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.9395 - val_accuracy: 0.8571 - val_loss: 0.9195\n",
"Epoch 267/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9419 - val_accuracy: 0.8571 - val_loss: 0.9182\n",
"Epoch 268/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9278 - val_accuracy: 0.8571 - val_loss: 0.9171\n",
"Epoch 269/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9086 - val_accuracy: 0.8571 - val_loss: 0.9159\n",
"Epoch 270/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9250 - val_accuracy: 0.8571 - val_loss: 0.9145\n",
"Epoch 271/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9115 - val_accuracy: 0.8571 - val_loss: 0.9132\n",
"Epoch 272/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9435 - val_accuracy: 0.8571 - val_loss: 0.9118\n",
"Epoch 273/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9561 - val_accuracy: 0.8571 - val_loss: 0.9108\n",
"Epoch 274/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9403 - val_accuracy: 0.8571 - val_loss: 0.9099\n",
"Epoch 275/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9595 - val_accuracy: 0.8571 - val_loss: 0.9086\n",
"Epoch 276/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9120 - val_accuracy: 0.8571 - val_loss: 0.9076\n",
"Epoch 277/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9327 - val_accuracy: 0.8571 - val_loss: 0.9065\n",
"Epoch 278/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9364 - val_accuracy: 0.8571 - val_loss: 0.9056\n",
"Epoch 279/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9078 - val_accuracy: 0.8571 - val_loss: 0.9044\n",
"Epoch 280/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8879 - val_accuracy: 0.8571 - val_loss: 0.9031\n",
"Epoch 281/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9402 - val_accuracy: 0.8571 - val_loss: 0.9023\n",
"Epoch 282/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9134 - val_accuracy: 0.8571 - val_loss: 0.9009\n",
"Epoch 283/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9364 - val_accuracy: 0.8571 - val_loss: 0.8997\n",
"Epoch 284/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8979 - val_accuracy: 0.8571 - val_loss: 0.8985\n",
"Epoch 285/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.9045 - val_accuracy: 0.8571 - val_loss: 0.8975\n",
"Epoch 286/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.9414 - val_accuracy: 0.8571 - val_loss: 0.8964\n",
"Epoch 287/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8925 - val_accuracy: 0.8571 - val_loss: 0.8951\n",
"Epoch 288/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.9091 - val_accuracy: 0.8571 - val_loss: 0.8944\n",
"Epoch 289/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9149 - val_accuracy: 0.8571 - val_loss: 0.8936\n",
"Epoch 290/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8909 - val_accuracy: 0.8571 - val_loss: 0.8925\n",
"Epoch 291/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8932 - val_accuracy: 0.8571 - val_loss: 0.8912\n",
"Epoch 292/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9086 - val_accuracy: 0.8571 - val_loss: 0.8901\n",
"Epoch 293/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9072 - val_accuracy: 0.8571 - val_loss: 0.8890\n",
"Epoch 294/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9204 - val_accuracy: 0.8571 - val_loss: 0.8879\n",
"Epoch 295/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8734 - val_accuracy: 0.8571 - val_loss: 0.8866\n",
"Epoch 296/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9355 - val_accuracy: 0.8571 - val_loss: 0.8856\n",
"Epoch 297/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8829 - val_accuracy: 0.8571 - val_loss: 0.8843\n",
"Epoch 298/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8882 - val_accuracy: 0.8571 - val_loss: 0.8831\n",
"Epoch 299/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9170 - val_accuracy: 0.8571 - val_loss: 0.8821\n",
"Epoch 300/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8941 - val_accuracy: 0.8571 - val_loss: 0.8812\n",
"Epoch 301/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8746 - val_accuracy: 0.8571 - val_loss: 0.8801\n",
"Epoch 302/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8817 - val_accuracy: 0.8571 - val_loss: 0.8789\n",
"Epoch 303/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9101 - val_accuracy: 0.8571 - val_loss: 0.8777\n",
"Epoch 304/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8715 - val_accuracy: 0.8571 - val_loss: 0.8766\n",
"Epoch 305/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.8919 - val_accuracy: 0.8571 - val_loss: 0.8756\n",
"Epoch 306/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9159 - val_accuracy: 0.8571 - val_loss: 0.8745\n",
"Epoch 307/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.9122 - val_accuracy: 0.8571 - val_loss: 0.8736\n",
"Epoch 308/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8950 - val_accuracy: 0.8571 - val_loss: 0.8726\n",
"Epoch 309/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8599 - val_accuracy: 0.8571 - val_loss: 0.8716\n",
"Epoch 310/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.9009 - val_accuracy: 0.8571 - val_loss: 0.8705\n",
"Epoch 311/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8738 - val_accuracy: 0.8571 - val_loss: 0.8694\n",
"Epoch 312/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8587 - val_accuracy: 0.8571 - val_loss: 0.8682\n",
"Epoch 313/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8846 - val_accuracy: 0.8571 - val_loss: 0.8674\n",
"Epoch 314/500\n",
"2/2 - 0s - 28ms/step - accuracy: 0.8537 - loss: 0.8935 - val_accuracy: 0.8571 - val_loss: 0.8664\n",
"Epoch 315/500\n",
"2/2 - 0s - 30ms/step - accuracy: 0.8537 - loss: 0.8979 - val_accuracy: 0.8571 - val_loss: 0.8655\n",
"Epoch 316/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8693 - val_accuracy: 0.8571 - val_loss: 0.8644\n",
"Epoch 317/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8792 - val_accuracy: 0.8571 - val_loss: 0.8633\n",
"Epoch 318/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8820 - val_accuracy: 0.8571 - val_loss: 0.8626\n",
"Epoch 319/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8652 - val_accuracy: 0.8571 - val_loss: 0.8614\n",
"Epoch 320/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8814 - val_accuracy: 0.8571 - val_loss: 0.8605\n",
"Epoch 321/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8621 - val_accuracy: 0.8571 - val_loss: 0.8594\n",
"Epoch 322/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8744 - val_accuracy: 0.8571 - val_loss: 0.8585\n",
"Epoch 323/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8805 - val_accuracy: 0.8571 - val_loss: 0.8575\n",
"Epoch 324/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8818 - val_accuracy: 0.8571 - val_loss: 0.8568\n",
"Epoch 325/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8989 - val_accuracy: 0.8571 - val_loss: 0.8560\n",
"Epoch 326/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8706 - val_accuracy: 0.8571 - val_loss: 0.8548\n",
"Epoch 327/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8623 - val_accuracy: 0.8571 - val_loss: 0.8539\n",
"Epoch 328/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8570 - val_accuracy: 0.8571 - val_loss: 0.8529\n",
"Epoch 329/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8858 - val_accuracy: 0.8571 - val_loss: 0.8519\n",
"Epoch 330/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8571 - val_accuracy: 0.8571 - val_loss: 0.8512\n",
"Epoch 331/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8839 - val_accuracy: 0.8571 - val_loss: 0.8503\n",
"Epoch 332/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8556 - val_accuracy: 0.8571 - val_loss: 0.8492\n",
"Epoch 333/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8764 - val_accuracy: 0.8571 - val_loss: 0.8482\n",
"Epoch 334/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8353 - val_accuracy: 0.8571 - val_loss: 0.8470\n",
"Epoch 335/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8659 - val_accuracy: 0.8571 - val_loss: 0.8461\n",
"Epoch 336/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8745 - val_accuracy: 0.8571 - val_loss: 0.8452\n",
"Epoch 337/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8465 - val_accuracy: 0.8571 - val_loss: 0.8444\n",
"Epoch 338/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8740 - val_accuracy: 0.8571 - val_loss: 0.8434\n",
"Epoch 339/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8760 - val_accuracy: 0.8571 - val_loss: 0.8423\n",
"Epoch 340/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8553 - val_accuracy: 0.8571 - val_loss: 0.8415\n",
"Epoch 341/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8460 - val_accuracy: 0.8571 - val_loss: 0.8406\n",
"Epoch 342/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8487 - val_accuracy: 0.8571 - val_loss: 0.8397\n",
"Epoch 343/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8667 - val_accuracy: 0.8571 - val_loss: 0.8387\n",
"Epoch 344/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8513 - val_accuracy: 0.8571 - val_loss: 0.8378\n",
"Epoch 345/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8692 - val_accuracy: 0.8571 - val_loss: 0.8371\n",
"Epoch 346/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8271 - val_accuracy: 0.8571 - val_loss: 0.8360\n",
"Epoch 347/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8738 - val_accuracy: 0.8571 - val_loss: 0.8353\n",
"Epoch 348/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8613 - val_accuracy: 0.8571 - val_loss: 0.8344\n",
"Epoch 349/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8322 - val_accuracy: 0.8571 - val_loss: 0.8334\n",
"Epoch 350/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8745 - val_accuracy: 0.8571 - val_loss: 0.8325\n",
"Epoch 351/500\n",
"2/2 - 0s - 26ms/step - accuracy: 0.8537 - loss: 0.8528 - val_accuracy: 0.8571 - val_loss: 0.8316\n",
"Epoch 352/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8418 - val_accuracy: 0.8571 - val_loss: 0.8306\n",
"Epoch 353/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8466 - val_accuracy: 0.8571 - val_loss: 0.8296\n",
"Epoch 354/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8299 - val_accuracy: 0.8571 - val_loss: 0.8287\n",
"Epoch 355/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8493 - val_accuracy: 0.8571 - val_loss: 0.8277\n",
"Epoch 356/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8468 - val_accuracy: 0.8571 - val_loss: 0.8267\n",
"Epoch 357/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8561 - val_accuracy: 0.8571 - val_loss: 0.8261\n",
"Epoch 358/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8319 - val_accuracy: 0.8571 - val_loss: 0.8252\n",
"Epoch 359/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8462 - val_accuracy: 0.8571 - val_loss: 0.8244\n",
"Epoch 360/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8342 - val_accuracy: 0.8571 - val_loss: 0.8235\n",
"Epoch 361/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8424 - val_accuracy: 0.8571 - val_loss: 0.8225\n",
"Epoch 362/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8315 - val_accuracy: 0.8571 - val_loss: 0.8216\n",
"Epoch 363/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8650 - val_accuracy: 0.8571 - val_loss: 0.8210\n",
"Epoch 364/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8644 - val_accuracy: 0.8571 - val_loss: 0.8201\n",
"Epoch 365/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8452 - val_accuracy: 0.8571 - val_loss: 0.8193\n",
"Epoch 366/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8463 - val_accuracy: 0.8571 - val_loss: 0.8186\n",
"Epoch 367/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8374 - val_accuracy: 0.8571 - val_loss: 0.8176\n",
"Epoch 368/500\n",
"2/2 - 0s - 17ms/step - accuracy: 0.8537 - loss: 0.8477 - val_accuracy: 0.8571 - val_loss: 0.8168\n",
"Epoch 369/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8179 - val_accuracy: 0.8571 - val_loss: 0.8159\n",
"Epoch 370/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7949 - val_accuracy: 0.8571 - val_loss: 0.8150\n",
"Epoch 371/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8133 - val_accuracy: 0.8571 - val_loss: 0.8141\n",
"Epoch 372/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8204 - val_accuracy: 0.8571 - val_loss: 0.8132\n",
"Epoch 373/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8575 - val_accuracy: 0.8571 - val_loss: 0.8124\n",
"Epoch 374/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8066 - val_accuracy: 0.8571 - val_loss: 0.8115\n",
"Epoch 375/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8284 - val_accuracy: 0.8571 - val_loss: 0.8106\n",
"Epoch 376/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8040 - val_accuracy: 0.8571 - val_loss: 0.8097\n",
"Epoch 377/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8243 - val_accuracy: 0.8571 - val_loss: 0.8091\n",
"Epoch 378/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8145 - val_accuracy: 0.8571 - val_loss: 0.8084\n",
"Epoch 379/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8263 - val_accuracy: 0.8571 - val_loss: 0.8075\n",
"Epoch 380/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8289 - val_accuracy: 0.8571 - val_loss: 0.8068\n",
"Epoch 381/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8228 - val_accuracy: 0.8571 - val_loss: 0.8059\n",
"Epoch 382/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8157 - val_accuracy: 0.8571 - val_loss: 0.8050\n",
"Epoch 383/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8155 - val_accuracy: 0.8571 - val_loss: 0.8045\n",
"Epoch 384/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8240 - val_accuracy: 0.8571 - val_loss: 0.8036\n",
"Epoch 385/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8237 - val_accuracy: 0.8571 - val_loss: 0.8027\n",
"Epoch 386/500\n",
"2/2 - 0s - 18ms/step - accuracy: 0.8537 - loss: 0.8163 - val_accuracy: 0.8571 - val_loss: 0.8020\n",
"Epoch 387/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8028 - val_accuracy: 0.8571 - val_loss: 0.8012\n",
"Epoch 388/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7948 - val_accuracy: 0.8571 - val_loss: 0.8004\n",
"Epoch 389/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7891 - val_accuracy: 0.8571 - val_loss: 0.7996\n",
"Epoch 390/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8254 - val_accuracy: 0.8571 - val_loss: 0.7987\n",
"Epoch 391/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8277 - val_accuracy: 0.8571 - val_loss: 0.7981\n",
"Epoch 392/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8004 - val_accuracy: 0.8571 - val_loss: 0.7973\n",
"Epoch 393/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8248 - val_accuracy: 0.8571 - val_loss: 0.7966\n",
"Epoch 394/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8204 - val_accuracy: 0.8571 - val_loss: 0.7958\n",
"Epoch 395/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8073 - val_accuracy: 0.8571 - val_loss: 0.7950\n",
"Epoch 396/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.7913 - val_accuracy: 0.8571 - val_loss: 0.7942\n",
"Epoch 397/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8046 - val_accuracy: 0.8571 - val_loss: 0.7934\n",
"Epoch 398/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8163 - val_accuracy: 0.8571 - val_loss: 0.7927\n",
"Epoch 399/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.7983 - val_accuracy: 0.8571 - val_loss: 0.7918\n",
"Epoch 400/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8062 - val_accuracy: 0.8571 - val_loss: 0.7912\n",
"Epoch 401/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.8047 - val_accuracy: 0.8571 - val_loss: 0.7904\n",
"Epoch 402/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.8145 - val_accuracy: 0.8571 - val_loss: 0.7898\n",
"Epoch 403/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8128 - val_accuracy: 0.8571 - val_loss: 0.7892\n",
"Epoch 404/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8163 - val_accuracy: 0.8571 - val_loss: 0.7885\n",
"Epoch 405/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8054 - val_accuracy: 0.8571 - val_loss: 0.7877\n",
"Epoch 406/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8282 - val_accuracy: 0.8571 - val_loss: 0.7870\n",
"Epoch 407/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7878 - val_accuracy: 0.8571 - val_loss: 0.7862\n",
"Epoch 408/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7720 - val_accuracy: 0.8571 - val_loss: 0.7854\n",
"Epoch 409/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.8026 - val_accuracy: 0.8571 - val_loss: 0.7848\n",
"Epoch 410/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7872 - val_accuracy: 0.8571 - val_loss: 0.7840\n",
"Epoch 411/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7907 - val_accuracy: 0.8571 - val_loss: 0.7833\n",
"Epoch 412/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8070 - val_accuracy: 0.8571 - val_loss: 0.7828\n",
"Epoch 413/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7948 - val_accuracy: 0.8571 - val_loss: 0.7821\n",
"Epoch 414/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8079 - val_accuracy: 0.8571 - val_loss: 0.7815\n",
"Epoch 415/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8118 - val_accuracy: 0.8571 - val_loss: 0.7811\n",
"Epoch 416/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7934 - val_accuracy: 0.8571 - val_loss: 0.7804\n",
"Epoch 417/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7853 - val_accuracy: 0.8571 - val_loss: 0.7797\n",
"Epoch 418/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7973 - val_accuracy: 0.8571 - val_loss: 0.7792\n",
"Epoch 419/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7896 - val_accuracy: 0.8571 - val_loss: 0.7785\n",
"Epoch 420/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8097 - val_accuracy: 0.8571 - val_loss: 0.7779\n",
"Epoch 421/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.8039 - val_accuracy: 0.8571 - val_loss: 0.7773\n",
"Epoch 422/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8143 - val_accuracy: 0.8571 - val_loss: 0.7769\n",
"Epoch 423/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7784 - val_accuracy: 0.8571 - val_loss: 0.7763\n",
"Epoch 424/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7868 - val_accuracy: 0.8571 - val_loss: 0.7756\n",
"Epoch 425/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7817 - val_accuracy: 0.8571 - val_loss: 0.7748\n",
"Epoch 426/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7935 - val_accuracy: 0.8571 - val_loss: 0.7741\n",
"Epoch 427/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7944 - val_accuracy: 0.8571 - val_loss: 0.7734\n",
"Epoch 428/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7773 - val_accuracy: 0.8571 - val_loss: 0.7727\n",
"Epoch 429/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.8049 - val_accuracy: 0.8571 - val_loss: 0.7720\n",
"Epoch 430/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.8097 - val_accuracy: 0.8571 - val_loss: 0.7714\n",
"Epoch 431/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7760 - val_accuracy: 0.8571 - val_loss: 0.7707\n",
"Epoch 432/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7626 - val_accuracy: 0.8571 - val_loss: 0.7701\n",
"Epoch 433/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.7663 - val_accuracy: 0.8571 - val_loss: 0.7694\n",
"Epoch 434/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.8070 - val_accuracy: 0.8571 - val_loss: 0.7687\n",
"Epoch 435/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7924 - val_accuracy: 0.8571 - val_loss: 0.7680\n",
"Epoch 436/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7837 - val_accuracy: 0.8571 - val_loss: 0.7674\n",
"Epoch 437/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.7776 - val_accuracy: 0.8571 - val_loss: 0.7668\n",
"Epoch 438/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7733 - val_accuracy: 0.8571 - val_loss: 0.7662\n",
"Epoch 439/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7763 - val_accuracy: 0.8571 - val_loss: 0.7655\n",
"Epoch 440/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7831 - val_accuracy: 0.8571 - val_loss: 0.7649\n",
"Epoch 441/500\n",
"2/2 - 0s - 18ms/step - accuracy: 0.8537 - loss: 0.7754 - val_accuracy: 0.8571 - val_loss: 0.7642\n",
"Epoch 442/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7793 - val_accuracy: 0.8571 - val_loss: 0.7636\n",
"Epoch 443/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7956 - val_accuracy: 0.8571 - val_loss: 0.7630\n",
"Epoch 444/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7908 - val_accuracy: 0.8571 - val_loss: 0.7623\n",
"Epoch 445/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7548 - val_accuracy: 0.8571 - val_loss: 0.7617\n",
"Epoch 446/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7658 - val_accuracy: 0.8571 - val_loss: 0.7610\n",
"Epoch 447/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7624 - val_accuracy: 0.8571 - val_loss: 0.7606\n",
"Epoch 448/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7858 - val_accuracy: 0.8571 - val_loss: 0.7600\n",
"Epoch 449/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7509 - val_accuracy: 0.8571 - val_loss: 0.7593\n",
"Epoch 450/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7449 - val_accuracy: 0.8571 - val_loss: 0.7585\n",
"Epoch 451/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7845 - val_accuracy: 0.8571 - val_loss: 0.7581\n",
"Epoch 452/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7731 - val_accuracy: 0.8571 - val_loss: 0.7575\n",
"Epoch 453/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7696 - val_accuracy: 0.8571 - val_loss: 0.7568\n",
"Epoch 454/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7762 - val_accuracy: 0.8571 - val_loss: 0.7562\n",
"Epoch 455/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7623 - val_accuracy: 0.8571 - val_loss: 0.7558\n",
"Epoch 456/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7681 - val_accuracy: 0.8571 - val_loss: 0.7552\n",
"Epoch 457/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.7775 - val_accuracy: 0.8571 - val_loss: 0.7546\n",
"Epoch 458/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7442 - val_accuracy: 0.8571 - val_loss: 0.7539\n",
"Epoch 459/500\n",
"2/2 - 0s - 17ms/step - accuracy: 0.8537 - loss: 0.7491 - val_accuracy: 0.8571 - val_loss: 0.7533\n",
"Epoch 460/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7569 - val_accuracy: 0.8571 - val_loss: 0.7526\n",
"Epoch 461/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7578 - val_accuracy: 0.8571 - val_loss: 0.7520\n",
"Epoch 462/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7868 - val_accuracy: 0.8571 - val_loss: 0.7514\n",
"Epoch 463/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7384 - val_accuracy: 0.8571 - val_loss: 0.7507\n",
"Epoch 464/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.7416 - val_accuracy: 0.8571 - val_loss: 0.7501\n",
"Epoch 465/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7469 - val_accuracy: 0.8571 - val_loss: 0.7495\n",
"Epoch 466/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7544 - val_accuracy: 0.8571 - val_loss: 0.7489\n",
"Epoch 467/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7512 - val_accuracy: 0.8571 - val_loss: 0.7483\n",
"Epoch 468/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7596 - val_accuracy: 0.8571 - val_loss: 0.7478\n",
"Epoch 469/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7349 - val_accuracy: 0.8571 - val_loss: 0.7473\n",
"Epoch 470/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7470 - val_accuracy: 0.8571 - val_loss: 0.7467\n",
"Epoch 471/500\n",
"2/2 - 0s - 16ms/step - accuracy: 0.8537 - loss: 0.7698 - val_accuracy: 0.8571 - val_loss: 0.7461\n",
"Epoch 472/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7768 - val_accuracy: 0.8571 - val_loss: 0.7455\n",
"Epoch 473/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7659 - val_accuracy: 0.8571 - val_loss: 0.7449\n",
"Epoch 474/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7718 - val_accuracy: 0.8571 - val_loss: 0.7443\n",
"Epoch 475/500\n",
"2/2 - 0s - 19ms/step - accuracy: 0.8537 - loss: 0.7533 - val_accuracy: 0.8571 - val_loss: 0.7437\n",
"Epoch 476/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7394 - val_accuracy: 0.8571 - val_loss: 0.7432\n",
"Epoch 477/500\n",
"2/2 - 0s - 23ms/step - accuracy: 0.8537 - loss: 0.7536 - val_accuracy: 0.8571 - val_loss: 0.7426\n",
"Epoch 478/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7421 - val_accuracy: 0.8571 - val_loss: 0.7421\n",
"Epoch 479/500\n",
"2/2 - 0s - 18ms/step - accuracy: 0.8537 - loss: 0.7407 - val_accuracy: 0.8571 - val_loss: 0.7414\n",
"Epoch 480/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7582 - val_accuracy: 0.8571 - val_loss: 0.7409\n",
"Epoch 481/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7463 - val_accuracy: 0.8571 - val_loss: 0.7403\n",
"Epoch 482/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7320 - val_accuracy: 0.8571 - val_loss: 0.7397\n",
"Epoch 483/500\n",
"2/2 - 0s - 25ms/step - accuracy: 0.8537 - loss: 0.7533 - val_accuracy: 0.8571 - val_loss: 0.7391\n",
"Epoch 484/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7584 - val_accuracy: 0.8571 - val_loss: 0.7385\n",
"Epoch 485/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7506 - val_accuracy: 0.8571 - val_loss: 0.7379\n",
"Epoch 486/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7710 - val_accuracy: 0.8571 - val_loss: 0.7375\n",
"Epoch 487/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.7334 - val_accuracy: 0.8571 - val_loss: 0.7369\n",
"Epoch 488/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7446 - val_accuracy: 0.8571 - val_loss: 0.7365\n",
"Epoch 489/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7415 - val_accuracy: 0.8571 - val_loss: 0.7360\n",
"Epoch 490/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7383 - val_accuracy: 0.8571 - val_loss: 0.7354\n",
"Epoch 491/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7500 - val_accuracy: 0.8571 - val_loss: 0.7348\n",
"Epoch 492/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7453 - val_accuracy: 0.8571 - val_loss: 0.7343\n",
"Epoch 493/500\n",
"2/2 - 0s - 22ms/step - accuracy: 0.8537 - loss: 0.7379 - val_accuracy: 0.8571 - val_loss: 0.7337\n",
"Epoch 494/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7510 - val_accuracy: 0.8571 - val_loss: 0.7333\n",
"Epoch 495/500\n",
"2/2 - 0s - 24ms/step - accuracy: 0.8537 - loss: 0.7632 - val_accuracy: 0.8571 - val_loss: 0.7327\n",
"Epoch 496/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7321 - val_accuracy: 0.8571 - val_loss: 0.7323\n",
"Epoch 497/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7605 - val_accuracy: 0.8571 - val_loss: 0.7317\n",
"Epoch 498/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7427 - val_accuracy: 0.8571 - val_loss: 0.7311\n",
"Epoch 499/500\n",
"2/2 - 0s - 20ms/step - accuracy: 0.8537 - loss: 0.7268 - val_accuracy: 0.8571 - val_loss: 0.7305\n",
"Epoch 500/500\n",
"2/2 - 0s - 21ms/step - accuracy: 0.8537 - loss: 0.7267 - val_accuracy: 0.8571 - val_loss: 0.7300\n",
"1/1 - 0s - 16ms/step - accuracy: 0.8571 - loss: 0.7300\n",
"Dokładność testowa: 85.71%\n"
]
}
],
"source": [
"history = model.fit(X_train, y_train, epochs=500, validation_data=(X_test, y_test), verbose=2)\n",
"\n",
"test_loss, test_accuracy = model.evaluate(X_test, y_test, verbose=2)\n",
"print(f\"Dokładność testowa: {test_accuracy:.2%}\")"
]
},
{
"cell_type": "markdown",
"id": "84409c40-7973-4e65-b81f-72d0837e8781",
"metadata": {},
"source": [
"## Efekty uczenia"
]
},
{
"cell_type": "markdown",
"id": "ef8129a5-7c32-4559-a717-c4737b9abe76",
"metadata": {},
"source": [
"Wytrenowany model osiąga skuteczność predykcji na poziomie 85,7%"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "8ddb08bb-0f04-420e-9749-8c7218bb729a",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step - accuracy: 0.8571 - loss: 0.7300\n"
]
},
{
"data": {
"text/plain": [
"0.8571428656578064"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.evaluate(X_test, y_test)[1]"
]
},
{
"cell_type": "markdown",
"id": "cb9ef13b-39d7-4526-be9c-15e165f8c50a",
"metadata": {},
"source": [
"## Wykresy"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "5f24c9b1-002c-43d3-962c-089671824e92",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(history.history['loss'])\n",
"plt.plot(history.history['val_loss'])\n",
"plt.title('Model loss')\n",
"plt.ylabel('Loss')\n",
"plt.xlabel('Epoch')\n",
"plt.legend(['Train', 'Val'], loc='upper right')\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "d6ff63d4-57ad-48a3-8104-bb4a47d14dcc",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.plot(history.history['accuracy'])\n",
"plt.plot(history.history['val_accuracy'])\n",
"plt.title('Model accuracy')\n",
"plt.ylabel('Accuracy')\n",
"plt.xlabel('Epoch')\n",
"plt.legend(['Train', 'Val'], loc='lower right')\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"id": "75484222-6113-4e1e-a0e9-d371355ba02b",
"metadata": {},
"source": [
"## Zapisanie modelu do pliku"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "df41cdc5-73f1-472e-bd5c-1eb5324849b6",
"metadata": {},
"outputs": [],
"source": [
"model.save('./model.keras') "
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "a0aa450e-81be-4987-8125-8a53556e83de",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.0"
}
},
"nbformat": 4,
"nbformat_minor": 5
}