cellphone.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a06d1b95-c763-4724-9caf-d955cabec0ec",
   "metadata": {},
   "source": [
    "## Predicting the price of your cell phone with a Neural Network! \n",
    "#### Have you ever wondered if you are paying too much for your cell phone? Let's train a model using multi-class classification to find out! "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "11691df1-4983-4d7a-834c-df03bd75d7fc",
   "metadata": {},
   "source": [
    "#### Notes\n",
    "- Dataset is taken from [Kaggle](https://www.kaggle.com/datasets/atefehmirnaseri/cell-phone-price)\n",
    "- This problem is a multi-class classification problem\n",
    "- We will be building a Nerual Network with tensorflow\n",
    "- This Neural Network will use the softmax algo and logistic loss function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "82fb99fa-3773-473a-97a7-01e4380a9c78",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras import Sequential\n",
    "from tensorflow.keras.layers import Dense\n",
    "from tensorflow.keras.losses import SparseCategoricalCrossentropy\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "import statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c8298496-31a0-4192-a119-6f651d6730bb",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train.shape: (2000, 20) Y_train.shape: (2000,) \n",
      "\n",
      "There are 2000 training set and 20 features in the dataset, the 20 features are: \n",
      "1: battery_power\n",
      "2: blue\n",
      "3: clock_speed\n",
      "4: dual_sim\n",
      "5: fc\n",
      "6: four_g\n",
      "7: int_memory\n",
      "8: m_dep\n",
      "9: mobile_wt\n",
      "10: n_cores\n",
      "11: pc\n",
      "12: px_height\n",
      "13: px_width\n",
      "14: ram\n",
      "15: sc_h\n",
      "16: sc_w\n",
      "17: talk_time\n",
      "18: three_g\n",
      "19: touch_screen\n",
      "20: wifi\n",
      "\n",
      " There are 4 distinct output values or classes\n"
     ]
    }
   ],
   "source": [
    "#Understand the data set \n",
    "df = pd.read_csv('CellPhone_train.csv')\n",
    "\n",
    "Y_train = np.array(df.pop('price_range'))\n",
    "X_train = np.array(df)\n",
    "X_features = df.columns.values\n",
    "\n",
    "m,n = X_train.shape\n",
    "\n",
    "print(f\"X_train.shape: {X_train.shape} Y_train.shape: {Y_train.shape} \\n\")\n",
    "print(f\"There are {m} training set and {n} features in the dataset, the {n} features are: \")\n",
    "for i in range(n):\n",
    "    print(f\"{i+1}: {X_features[i]}\")\n",
    "\n",
    "print(f\"\\n There are {len(np.unique(Y_train))} distinct output values or classes\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "472ff2a9-ba4a-447c-b2d0-9d2943e9eb52",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">18</span>)             │           <span style=\"color: #00af00; text-decoration-color: #00af00\">378</span> │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">4</span>)              │            <span style=\"color: #00af00; text-decoration-color: #00af00\">76</span> │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m18\u001b[0m)             │           \u001b[38;5;34m378\u001b[0m │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m4\u001b[0m)              │            \u001b[38;5;34m76\u001b[0m │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">454</span> (1.77 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m454\u001b[0m (1.77 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">454</span> (1.77 KB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m454\u001b[0m (1.77 KB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#building the neural network\n",
    "model = Sequential([\n",
    "    tf.keras.Input(shape=(20,)),\n",
    "    Dense(units=18, activation='relu'),\n",
    "    Dense(units=4, activation='linear'),\n",
    "    ])\n",
    "model.compile(loss=SparseCategoricalCrossentropy(from_logits=True), optimizer=Adam(learning_rate=1e-2))\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5bcbf7ba-b6d5-4e7e-a65b-a55c04c4a39b",
   "metadata": {},
   "source": [
    "#### Structure\n",
    "The Nerual Network will have ***3*** layers\n",
    "- First Layer: Input layer and will take the shape of the input parameter which has **20** different features\n",
    "    > will have shape $(m, 20)$ where $m$ is the number of training set.\n",
    "- Second Layer: Hidden layer and will have **18** neurons. Total # of params is $20*18(w) + 18(b) = 378$\n",
    "    > will have output of shape $(m, 18)$ where $m$ is the number of training set.\n",
    "- Output Layer: Since it is a classification problem with 4 outputs, it will have **4** neurons. Total # of params is $18*4(w) + 4 = 76$\n",
    "    > will have output shape $(m, 4)$ where $m$ is the number of training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a9a4f96c-0df8-4961-97d3-93bb924806ef",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 253us/step - loss: 189.6533\n",
      "Epoch 2/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 7.9861\n",
      "Epoch 3/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 3.5992\n",
      "Epoch 4/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 3.7222\n",
      "Epoch 5/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 7.6736\n",
      "Epoch 6/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 5.6260\n",
      "Epoch 7/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 3.4080\n",
      "Epoch 8/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 2.3414\n",
      "Epoch 9/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 2.1996\n",
      "Epoch 10/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 3.7292\n",
      "Epoch 11/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 191us/step - loss: 2.1246\n",
      "Epoch 12/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 1.9319\n",
      "Epoch 13/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 4.0779\n",
      "Epoch 14/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 191us/step - loss: 2.0092\n",
      "Epoch 15/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 1.7011\n",
      "Epoch 16/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 4.0670\n",
      "Epoch 17/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 546us/step - loss: 2.4764\n",
      "Epoch 18/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 2.8017\n",
      "Epoch 19/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 2.4078\n",
      "Epoch 20/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 1.3441\n",
      "Epoch 21/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 3.8642\n",
      "Epoch 22/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 1.5220\n",
      "Epoch 23/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 2.2716\n",
      "Epoch 24/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 2.0383\n",
      "Epoch 25/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 1.4918\n",
      "Epoch 26/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 2.0677\n",
      "Epoch 27/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 1.4634\n",
      "Epoch 28/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 550us/step - loss: 1.3120\n",
      "Epoch 29/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 4.0195\n",
      "Epoch 30/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 1.7691\n",
      "Epoch 31/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 1.6790\n",
      "Epoch 32/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 1.1663\n",
      "Epoch 33/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 1.3594\n",
      "Epoch 34/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.9688\n",
      "Epoch 35/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 1.5361\n",
      "Epoch 36/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 220us/step - loss: 1.3327\n",
      "Epoch 37/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.9953\n",
      "Epoch 38/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 1.4770\n",
      "Epoch 39/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 1.0912\n",
      "Epoch 40/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 1.4668\n",
      "Epoch 41/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 2.1158\n",
      "Epoch 42/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.7761\n",
      "Epoch 43/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.7738\n",
      "Epoch 44/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.8397\n",
      "Epoch 45/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.9156\n",
      "Epoch 46/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 1.0238\n",
      "Epoch 47/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 267us/step - loss: 0.7469\n",
      "Epoch 48/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.9825\n",
      "Epoch 49/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.7887\n",
      "Epoch 50/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.8930\n",
      "Epoch 51/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.9678\n",
      "Epoch 52/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.6745\n",
      "Epoch 53/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.9740\n",
      "Epoch 54/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 221us/step - loss: 0.7083\n",
      "Epoch 55/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.8340\n",
      "Epoch 56/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.6616\n",
      "Epoch 57/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.7413\n",
      "Epoch 58/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.7375\n",
      "Epoch 59/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.7953\n",
      "Epoch 60/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.5767\n",
      "Epoch 61/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.5455\n",
      "Epoch 62/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 564us/step - loss: 0.6182\n",
      "Epoch 63/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.6207\n",
      "Epoch 64/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.5084\n",
      "Epoch 65/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.5960\n",
      "Epoch 66/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4963\n",
      "Epoch 67/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.5358\n",
      "Epoch 68/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.5112\n",
      "Epoch 69/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.5088\n",
      "Epoch 70/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4942\n",
      "Epoch 71/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.5156\n",
      "Epoch 72/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.5428\n",
      "Epoch 73/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.5095\n",
      "Epoch 74/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 216us/step - loss: 0.4928\n",
      "Epoch 75/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 587us/step - loss: 0.4930\n",
      "Epoch 76/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.5349\n",
      "Epoch 77/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4917\n",
      "Epoch 78/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 221us/step - loss: 0.5185\n",
      "Epoch 79/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.4520\n",
      "Epoch 80/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 224us/step - loss: 0.4941\n",
      "Epoch 81/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.5062\n",
      "Epoch 82/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 218us/step - loss: 0.4670\n",
      "Epoch 83/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 264us/step - loss: 0.4427\n",
      "Epoch 84/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.5052\n",
      "Epoch 85/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4578\n",
      "Epoch 86/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 277us/step - loss: 0.4710\n",
      "Epoch 87/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.4570\n",
      "Epoch 88/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.4996\n",
      "Epoch 89/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4766\n",
      "Epoch 90/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.4677\n",
      "Epoch 91/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4552\n",
      "Epoch 92/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.4382\n",
      "Epoch 93/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4593\n",
      "Epoch 94/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.4521\n",
      "Epoch 95/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.4484\n",
      "Epoch 96/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 573us/step - loss: 0.4379\n",
      "Epoch 97/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4437\n",
      "Epoch 98/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.4601\n",
      "Epoch 99/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.4500\n",
      "Epoch 100/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.5043\n",
      "Epoch 101/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.4565\n",
      "Epoch 102/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.4399\n",
      "Epoch 103/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.4544\n",
      "Epoch 104/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.4218\n",
      "Epoch 105/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.4119\n",
      "Epoch 106/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 290us/step - loss: 0.4401\n",
      "Epoch 107/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.4432\n",
      "Epoch 108/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 220us/step - loss: 0.4513\n",
      "Epoch 109/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 216us/step - loss: 0.4504\n",
      "Epoch 110/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4773\n",
      "Epoch 111/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.4419\n",
      "Epoch 112/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 215us/step - loss: 0.4285\n",
      "Epoch 113/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.4505\n",
      "Epoch 114/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.4787\n",
      "Epoch 115/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 390us/step - loss: 0.4535\n",
      "Epoch 116/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 226us/step - loss: 0.4933\n",
      "Epoch 117/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.4573\n",
      "Epoch 118/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4371\n",
      "Epoch 119/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 220us/step - loss: 0.4549\n",
      "Epoch 120/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.4328\n",
      "Epoch 121/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.4111\n",
      "Epoch 122/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.4232\n",
      "Epoch 123/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 568us/step - loss: 0.4353\n",
      "Epoch 124/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.4296\n",
      "Epoch 125/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.5145\n",
      "Epoch 126/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.4445\n",
      "Epoch 127/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.4388\n",
      "Epoch 128/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.5580\n",
      "Epoch 129/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.4343\n",
      "Epoch 130/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4116\n",
      "Epoch 131/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 568us/step - loss: 0.4271\n",
      "Epoch 132/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.4077\n",
      "Epoch 133/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.4102\n",
      "Epoch 134/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.4461\n",
      "Epoch 135/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.4094\n",
      "Epoch 136/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4361\n",
      "Epoch 137/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.5203\n",
      "Epoch 138/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4220\n",
      "Epoch 139/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 276us/step - loss: 0.4025\n",
      "Epoch 140/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4458\n",
      "Epoch 141/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.4189\n",
      "Epoch 142/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.4506\n",
      "Epoch 143/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3855\n",
      "Epoch 144/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.4088\n",
      "Epoch 145/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4635\n",
      "Epoch 146/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 453us/step - loss: 0.4228\n",
      "Epoch 147/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.4922\n",
      "Epoch 148/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.4117\n",
      "Epoch 149/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.4381\n",
      "Epoch 150/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4313\n",
      "Epoch 151/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.4304\n",
      "Epoch 152/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.5194\n",
      "Epoch 153/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4188\n",
      "Epoch 154/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.5009\n",
      "Epoch 155/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.4369\n",
      "Epoch 156/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 218us/step - loss: 0.4737\n",
      "Epoch 157/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 221us/step - loss: 0.4188\n",
      "Epoch 158/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4327\n",
      "Epoch 159/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 222us/step - loss: 0.3996\n",
      "Epoch 160/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.3984\n",
      "Epoch 161/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4581\n",
      "Epoch 162/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.4282\n",
      "Epoch 163/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3992\n",
      "Epoch 164/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3906\n",
      "Epoch 165/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4004\n",
      "Epoch 166/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 237us/step - loss: 0.4737\n",
      "Epoch 167/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3672\n",
      "Epoch 168/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4071\n",
      "Epoch 169/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.4618\n",
      "Epoch 170/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.4472\n",
      "Epoch 171/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4207\n",
      "Epoch 172/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3804\n",
      "Epoch 173/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 552us/step - loss: 0.4428\n",
      "Epoch 174/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 216us/step - loss: 0.4445\n",
      "Epoch 175/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3993\n",
      "Epoch 176/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.4279\n",
      "Epoch 177/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.4213\n",
      "Epoch 178/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 216us/step - loss: 0.4086\n",
      "Epoch 179/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3481\n",
      "Epoch 180/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3935\n",
      "Epoch 181/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 572us/step - loss: 0.4812\n",
      "Epoch 182/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 293us/step - loss: 0.4430\n",
      "Epoch 183/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.4602\n",
      "Epoch 184/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.4569\n",
      "Epoch 185/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3841\n",
      "Epoch 186/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4131\n",
      "Epoch 187/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.4001\n",
      "Epoch 188/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3754\n",
      "Epoch 189/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.3811\n",
      "Epoch 190/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 558us/step - loss: 0.4234\n",
      "Epoch 191/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.3676\n",
      "Epoch 192/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 266us/step - loss: 0.3964\n",
      "Epoch 193/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.3994\n",
      "Epoch 194/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3963\n",
      "Epoch 195/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4015\n",
      "Epoch 196/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3566\n",
      "Epoch 197/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4146\n",
      "Epoch 198/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.4533\n",
      "Epoch 199/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3809\n",
      "Epoch 200/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 627us/step - loss: 0.4129\n",
      "Epoch 201/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.5071\n",
      "Epoch 202/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.4027\n",
      "Epoch 203/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.3622\n",
      "Epoch 204/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4791\n",
      "Epoch 205/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3573\n",
      "Epoch 206/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4118\n",
      "Epoch 207/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 612us/step - loss: 0.4275\n",
      "Epoch 208/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3682\n",
      "Epoch 209/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.4819\n",
      "Epoch 210/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.4382\n",
      "Epoch 211/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3536\n",
      "Epoch 212/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3464\n",
      "Epoch 213/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 191us/step - loss: 0.4250\n",
      "Epoch 214/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 287us/step - loss: 0.3639\n",
      "Epoch 215/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3859\n",
      "Epoch 216/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 191us/step - loss: 0.3695\n",
      "Epoch 217/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.4196\n",
      "Epoch 218/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.4541\n",
      "Epoch 219/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3525\n",
      "Epoch 220/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3563\n",
      "Epoch 221/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 272us/step - loss: 0.4045\n",
      "Epoch 222/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.4403\n",
      "Epoch 223/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.3599\n",
      "Epoch 224/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3542\n",
      "Epoch 225/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3684\n",
      "Epoch 226/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3655\n",
      "Epoch 227/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 547us/step - loss: 0.3496\n",
      "Epoch 228/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.4598\n",
      "Epoch 229/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3675\n",
      "Epoch 230/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.3678\n",
      "Epoch 231/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4786\n",
      "Epoch 232/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3858\n",
      "Epoch 233/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3368\n",
      "Epoch 234/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 559us/step - loss: 0.3890\n",
      "Epoch 235/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.3429\n",
      "Epoch 236/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3653\n",
      "Epoch 237/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.3374\n",
      "Epoch 238/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.4715\n",
      "Epoch 239/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.4367\n",
      "Epoch 240/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3965\n",
      "Epoch 241/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 589us/step - loss: 0.3152\n",
      "Epoch 242/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.4334\n",
      "Epoch 243/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4030\n",
      "Epoch 244/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3435\n",
      "Epoch 245/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.4368\n",
      "Epoch 246/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4649\n",
      "Epoch 247/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4262\n",
      "Epoch 248/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 635us/step - loss: 0.3453\n",
      "Epoch 249/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3371\n",
      "Epoch 250/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.3345\n",
      "Epoch 251/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3616\n",
      "Epoch 252/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3777\n",
      "Epoch 253/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.3463\n",
      "Epoch 254/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 610us/step - loss: 0.3277\n",
      "Epoch 255/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.3546\n",
      "Epoch 256/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3890\n",
      "Epoch 257/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3320\n",
      "Epoch 258/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.4805\n",
      "Epoch 259/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3673\n",
      "Epoch 260/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3234\n",
      "Epoch 261/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 630us/step - loss: 0.3816\n",
      "Epoch 262/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3810\n",
      "Epoch 263/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.3327\n",
      "Epoch 264/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.3537\n",
      "Epoch 265/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3476\n",
      "Epoch 266/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3505\n",
      "Epoch 267/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.4483\n",
      "Epoch 268/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 572us/step - loss: 0.3315\n",
      "Epoch 269/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.3277\n",
      "Epoch 270/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3326\n",
      "Epoch 271/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3432\n",
      "Epoch 272/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3731\n",
      "Epoch 273/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3345\n",
      "Epoch 274/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3329\n",
      "Epoch 275/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 576us/step - loss: 0.3417\n",
      "Epoch 276/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.2995\n",
      "Epoch 277/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.3322\n",
      "Epoch 278/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3898\n",
      "Epoch 279/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3489\n",
      "Epoch 280/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4103\n",
      "Epoch 281/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3310\n",
      "Epoch 282/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 645us/step - loss: 0.3507\n",
      "Epoch 283/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.3870\n",
      "Epoch 284/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3624\n",
      "Epoch 285/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3206\n",
      "Epoch 286/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3811\n",
      "Epoch 287/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3412\n",
      "Epoch 288/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3452\n",
      "Epoch 289/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 259us/step - loss: 0.3440\n",
      "Epoch 290/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3639\n",
      "Epoch 291/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3984\n",
      "Epoch 292/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4477\n",
      "Epoch 293/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3081\n",
      "Epoch 294/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3106\n",
      "Epoch 295/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 574us/step - loss: 0.3040\n",
      "Epoch 296/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 218us/step - loss: 0.3537\n",
      "Epoch 297/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3227\n",
      "Epoch 298/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 192us/step - loss: 0.3335\n",
      "Epoch 299/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3494\n",
      "Epoch 300/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.3011\n",
      "Epoch 301/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3298\n",
      "Epoch 302/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 551us/step - loss: 0.3606\n",
      "Epoch 303/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.3705\n",
      "Epoch 304/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.3782\n",
      "Epoch 305/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.4093\n",
      "Epoch 306/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3495\n",
      "Epoch 307/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3179\n",
      "Epoch 308/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3206\n",
      "Epoch 309/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 642us/step - loss: 0.3093\n",
      "Epoch 310/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.3279\n",
      "Epoch 311/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 205us/step - loss: 0.3261\n",
      "Epoch 312/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3058\n",
      "Epoch 313/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3360\n",
      "Epoch 314/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3239\n",
      "Epoch 315/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3003\n",
      "Epoch 316/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 643us/step - loss: 0.3187\n",
      "Epoch 317/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4024\n",
      "Epoch 318/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3427\n",
      "Epoch 319/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4949\n",
      "Epoch 320/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 190us/step - loss: 0.3294\n",
      "Epoch 321/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3101\n",
      "Epoch 322/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 214us/step - loss: 0.3236\n",
      "Epoch 323/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 292us/step - loss: 0.2845\n",
      "Epoch 324/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3213\n",
      "Epoch 325/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3530\n",
      "Epoch 326/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3377\n",
      "Epoch 327/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.3239\n",
      "Epoch 328/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3786\n",
      "Epoch 329/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3264\n",
      "Epoch 330/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 620us/step - loss: 0.3280\n",
      "Epoch 331/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3184\n",
      "Epoch 332/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3053\n",
      "Epoch 333/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3112\n",
      "Epoch 334/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3120\n",
      "Epoch 335/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3110\n",
      "Epoch 336/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3283\n",
      "Epoch 337/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 651us/step - loss: 0.4004\n",
      "Epoch 338/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3401\n",
      "Epoch 339/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.3890\n",
      "Epoch 340/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.4221\n",
      "Epoch 341/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 192us/step - loss: 0.3113\n",
      "Epoch 342/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3414\n",
      "Epoch 343/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3484\n",
      "Epoch 344/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 590us/step - loss: 0.3347\n",
      "Epoch 345/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.3277\n",
      "Epoch 346/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3901\n",
      "Epoch 347/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2873\n",
      "Epoch 348/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.2910\n",
      "Epoch 349/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3174\n",
      "Epoch 350/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2931\n",
      "Epoch 351/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 212us/step - loss: 0.3055\n",
      "Epoch 352/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 294us/step - loss: 0.2717\n",
      "Epoch 353/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3558\n",
      "Epoch 354/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3258\n",
      "Epoch 355/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.3003\n",
      "Epoch 356/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3270\n",
      "Epoch 357/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.3304\n",
      "Epoch 358/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3197\n",
      "Epoch 359/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 644us/step - loss: 0.3182\n",
      "Epoch 360/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3057\n",
      "Epoch 361/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.4353\n",
      "Epoch 362/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3595\n",
      "Epoch 363/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3042\n",
      "Epoch 364/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3122\n",
      "Epoch 365/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.2922\n",
      "Epoch 366/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 555us/step - loss: 0.3059\n",
      "Epoch 367/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 283us/step - loss: 0.3249\n",
      "Epoch 368/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.2963\n",
      "Epoch 369/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 192us/step - loss: 0.2944\n",
      "Epoch 370/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3195\n",
      "Epoch 371/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3474\n",
      "Epoch 372/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 209us/step - loss: 0.3269\n",
      "Epoch 373/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3305\n",
      "Epoch 374/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 651us/step - loss: 0.3110\n",
      "Epoch 375/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.2752\n",
      "Epoch 376/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2910\n",
      "Epoch 377/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.3086\n",
      "Epoch 378/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2900\n",
      "Epoch 379/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3128\n",
      "Epoch 380/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3841\n",
      "Epoch 381/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 567us/step - loss: 0.3075\n",
      "Epoch 382/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.2891\n",
      "Epoch 383/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3919\n",
      "Epoch 384/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3007\n",
      "Epoch 385/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3488\n",
      "Epoch 386/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.4248\n",
      "Epoch 387/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.2623\n",
      "Epoch 388/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.2749\n",
      "Epoch 389/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 642us/step - loss: 0.2728\n",
      "Epoch 390/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3365\n",
      "Epoch 391/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3399\n",
      "Epoch 392/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 207us/step - loss: 0.3431\n",
      "Epoch 393/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3082\n",
      "Epoch 394/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3020\n",
      "Epoch 395/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.3541\n",
      "Epoch 396/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 563us/step - loss: 0.2824\n",
      "Epoch 397/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 227us/step - loss: 0.3682\n",
      "Epoch 398/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2721\n",
      "Epoch 399/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3693\n",
      "Epoch 400/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3176\n",
      "Epoch 401/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2778\n",
      "Epoch 402/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.2787\n",
      "Epoch 403/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.4581\n",
      "Epoch 404/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 615us/step - loss: 0.2655\n",
      "Epoch 405/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2792\n",
      "Epoch 406/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3246\n",
      "Epoch 407/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3377\n",
      "Epoch 408/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3144\n",
      "Epoch 409/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3420\n",
      "Epoch 410/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2658\n",
      "Epoch 411/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3169\n",
      "Epoch 412/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 610us/step - loss: 0.2760\n",
      "Epoch 413/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2841\n",
      "Epoch 414/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 213us/step - loss: 0.2824\n",
      "Epoch 415/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3036\n",
      "Epoch 416/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2616\n",
      "Epoch 417/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2752\n",
      "Epoch 418/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.3178\n",
      "Epoch 419/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 190us/step - loss: 0.3064\n",
      "Epoch 420/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 296us/step - loss: 0.3269\n",
      "Epoch 421/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2842\n",
      "Epoch 422/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.2916\n",
      "Epoch 423/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.4570\n",
      "Epoch 424/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.3146\n",
      "Epoch 425/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.2804\n",
      "Epoch 426/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.3596\n",
      "Epoch 427/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3110\n",
      "Epoch 428/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 297us/step - loss: 0.3951\n",
      "Epoch 429/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.2897\n",
      "Epoch 430/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3334\n",
      "Epoch 431/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 216us/step - loss: 0.2547\n",
      "Epoch 432/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2548\n",
      "Epoch 433/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3427\n",
      "Epoch 434/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2789\n",
      "Epoch 435/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 582us/step - loss: 0.3260\n",
      "Epoch 436/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.2598\n",
      "Epoch 437/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3221\n",
      "Epoch 438/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3108\n",
      "Epoch 439/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 192us/step - loss: 0.2566\n",
      "Epoch 440/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2597\n",
      "Epoch 441/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.3108\n",
      "Epoch 442/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.3656\n",
      "Epoch 443/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 492us/step - loss: 0.3187\n",
      "Epoch 444/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.4023\n",
      "Epoch 445/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2929\n",
      "Epoch 446/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 221us/step - loss: 0.2489\n",
      "Epoch 447/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.3592\n",
      "Epoch 448/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3097\n",
      "Epoch 449/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.4151\n",
      "Epoch 450/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 545us/step - loss: 0.2677\n",
      "Epoch 451/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 217us/step - loss: 0.2469\n",
      "Epoch 452/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.3101\n",
      "Epoch 453/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.4067\n",
      "Epoch 454/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2856\n",
      "Epoch 455/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3025\n",
      "Epoch 456/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2682\n",
      "Epoch 457/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.2935\n",
      "Epoch 458/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 610us/step - loss: 0.2389\n",
      "Epoch 459/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.2679\n",
      "Epoch 460/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 204us/step - loss: 0.3306\n",
      "Epoch 461/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 203us/step - loss: 0.2960\n",
      "Epoch 462/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 219us/step - loss: 0.2662\n",
      "Epoch 463/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2581\n",
      "Epoch 464/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 206us/step - loss: 0.2661\n",
      "Epoch 465/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.5221\n",
      "Epoch 466/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 548us/step - loss: 0.2686\n",
      "Epoch 467/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2598\n",
      "Epoch 468/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.2987\n",
      "Epoch 469/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 200us/step - loss: 0.2967\n",
      "Epoch 470/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.2725\n",
      "Epoch 471/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 195us/step - loss: 0.2740\n",
      "Epoch 472/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2728\n",
      "Epoch 473/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 568us/step - loss: 0.2530\n",
      "Epoch 474/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 272us/step - loss: 0.2623\n",
      "Epoch 475/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.4370\n",
      "Epoch 476/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2642\n",
      "Epoch 477/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 199us/step - loss: 0.3010\n",
      "Epoch 478/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.2668\n",
      "Epoch 479/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 208us/step - loss: 0.2925\n",
      "Epoch 480/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2906\n",
      "Epoch 481/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 556us/step - loss: 0.2801\n",
      "Epoch 482/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 254us/step - loss: 0.3421\n",
      "Epoch 483/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2966\n",
      "Epoch 484/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.2979\n",
      "Epoch 485/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.3883\n",
      "Epoch 486/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 192us/step - loss: 0.2827\n",
      "Epoch 487/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 196us/step - loss: 0.3570\n",
      "Epoch 488/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.2406\n",
      "Epoch 489/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 589us/step - loss: 0.2709\n",
      "Epoch 490/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 220us/step - loss: 0.2555\n",
      "Epoch 491/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 193us/step - loss: 0.2424\n",
      "Epoch 492/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.3071\n",
      "Epoch 493/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 198us/step - loss: 0.2249\n",
      "Epoch 494/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 210us/step - loss: 0.2789\n",
      "Epoch 495/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 201us/step - loss: 0.2468\n",
      "Epoch 496/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 194us/step - loss: 0.3293\n",
      "Epoch 497/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 638us/step - loss: 0.2891\n",
      "Epoch 498/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 202us/step - loss: 0.2340\n",
      "Epoch 499/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 211us/step - loss: 0.2728\n",
      "Epoch 500/500\n",
      "\u001b[1m63/63\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 197us/step - loss: 0.2677\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<keras.src.callbacks.history.History at 0x167a037a0>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#train the model\n",
    "model.fit(X_train, Y_train, epochs=500)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00a82d48-f211-4207-a90e-5f4d2bef7e96",
   "metadata": {},
   "source": [
    "- Now that the model is trained, we can use the original dataset, $X\\_train$ to calculate the predicted results, $yhat$.\n",
    "- Let's then compare $yhat$ with $Y\\_train$ to calculate the accuracy of the model. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "47df288e-850a-4401-a3b4-99e71c9dd040",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy is 0.836\n"
     ]
    }
   ],
   "source": [
    "#predict and view accuracy \n",
    "logits = model(X_train)\n",
    "\n",
    "#Here, we are using softmax algo as the activation function\n",
    "f_x=tf.nn.softmax(logits).numpy() #shape of f_x is (2000, 4)\n",
    "\n",
    "yhat = []\n",
    "\n",
    "#softmax activation returns us with a list of probabilities of the respective indices being the \"correct\" value\n",
    "#we need to find out what is the highest probability within that list and return the indices\n",
    "for i in range(m):\n",
    "    yhat.append(np.argmax(f_x[i])) #np.argmax returns the indices of element with the highest numerical value within a list \n",
    "\n",
    "print(f\"Accuracy is {np.count_nonzero(yhat==Y_train)/m}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1ac33653-0e99-457d-a443-0e0bc6084b6c",
   "metadata": {},
   "source": [
    "#### Hooray! \n",
    "We managed to achieve accuracy of >0.80. Let's use our model to predict the price categories of a random set of data given in CellPhone_test.csv"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e1bf650f-5e23-4c0d-882a-d1e12b07b4f4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001b[1m32/32\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 398us/step\n",
      "[2, 3, 2, 3, 1, 2, 3, 1, 3, 0, 3, 3, 0, 0, 2, 0, 2, 1, 3, 2]\n"
     ]
    }
   ],
   "source": [
    "test  = pd.read_csv(\"CellPhone_test.csv\")\n",
    "\n",
    "#drop the id column\n",
    "test = test.drop(labels=['id'], axis=1)\n",
    "test = test.reindex(labels=X_features, axis=1)\n",
    "\n",
    "#convert pandas dataframe to numpy array\n",
    "test.to_numpy()\n",
    "\n",
    "#plug the data into the model\n",
    "prediction = model.predict(test)\n",
    "function=tf.nn.softmax(prediction).numpy()\n",
    "\n",
    "#softmax function\n",
    "results = [np.argmax(function[row]) for row in range(20)]\n",
    "\n",
    "print(results)"
   ]
  }
 ],
 "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": "3.12.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}