{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Introducing Keras\n",
"\n",
"Be sure to be using tensorflow 1.9 or newer!\n",
"\n",
"Keras is a higher-level API within TensorFlow that makes things a lot easier. Not only is it easier to use, it's easier to tune.\n",
"\n",
"Let's set up the same deep neural network we set up with TensorFlow to learn from the MNIST data set.\n",
"\n",
"First we'll import all the stuff we need, which will initialize Keras as a side effect:"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"from tensorflow import keras\n",
"from tensorflow.keras.datasets import mnist\n",
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Dense, Dropout\n",
"from tensorflow.keras.optimizers import RMSprop"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We'll load up the MNIST data set. Again, there are 60K training samples and 10K test samples."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [],
"source": [
"(mnist_train_images, mnist_train_labels), (mnist_test_images, mnist_test_labels) = mnist.load_data()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We need to explicitly convert the data into the format Keras / TensorFlow expects. We divide the image data by 255 in order to normalize it into 0-1 range, after converting it into floating point values."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"train_images = mnist_train_images.reshape(60000, 784)\n",
"test_images = mnist_test_images.reshape(10000, 784)\n",
"train_images = train_images.astype('float32')\n",
"test_images = test_images.astype('float32')\n",
"train_images /= 255\n",
"test_images /= 255"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we'll convert the 0-9 labels into \"one-hot\" format. Think of one_hot as a binary representation of the label data - that is, which number each handwriting sample was intended to represent. Mathematically one_hot represents a dimension for every possible label value. Every dimension is set to the value 0, except for the \"correct\" one which is set to 1. For example, the label vector representing the number 1 would be [0, 1, 0, 0, 0, 0, 0, 0, 0, 0] (remember we start counting at 0.) It's just a format that's optimized for how the labels are applied during training.\n",
"\n",
"So the training label data is a tensor of shape [60,000, 10] - 60,000 test images each associated with 10 binary values that indicate whether or not the image represents a given number from 0-9."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"train_labels = keras.utils.to_categorical(mnist_train_labels, 10)\n",
"test_labels = keras.utils.to_categorical(mnist_test_labels, 10)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Let's take a peek at one of the training images just to make sure it looks OK:"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[0. 0. 0. 1. 0. 0. 0. 0. 0. 0.]\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"def display_sample(num):\n",
" #Print the one-hot array of this sample's label \n",
" print(train_labels[num]) \n",
" #Print the label converted back to a number\n",
" label = train_labels[num].argmax(axis=0)\n",
" #Reshape the 768 values to a 28x28 image\n",
" image = train_images[num].reshape([28,28])\n",
" plt.title('Sample: %d Label: %d' % (num, label))\n",
" plt.imshow(image, cmap=plt.get_cmap('gray_r'))\n",
" plt.show()\n",
" \n",
"display_sample(1234)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here's where things get exciting. All that code we wrote in Tensorflow creating placeholders, variables, and defining a bunch of linear algebra for each layer in our neural network? None of that is necessary with Keras!\n",
"\n",
"We can set up the same layers like this. The input layer of 784 features feeds into a ReLU layer of 512 nodes, which then goes into 10 nodes with softmax applied. Couldn't be simpler:"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"from tensorflow.keras.layers import Input\n",
"model = Sequential()\n",
"model.add(Input(shape=(784,)))\n",
"model.add(Dense(512, activation='relu'))\n",
"model.add(Dense(10, activation='softmax'))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We can even get a nice description of the resulting model:"
]
},
{
"cell_type": "code",
"execution_count": 14,
"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\">512</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">401,920</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\">10</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">5,130</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;34m512\u001b[0m) │ \u001b[38;5;34m401,920\u001b[0m │\n",
"├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
"│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m10\u001b[0m) │ \u001b[38;5;34m5,130\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\">407,050</span> (1.55 MB)\n",
"</pre>\n"
],
"text/plain": [
"\u001b[1m Total params: \u001b[0m\u001b[38;5;34m407,050\u001b[0m (1.55 MB)\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\">407,050</span> (1.55 MB)\n",
"</pre>\n"
],
"text/plain": [
"\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m407,050\u001b[0m (1.55 MB)\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": [
"model.summary()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Setting up our optimizer and loss function is just as simple. We will use the RMSProp optimizer here. Other choices include Adagrad, SGD, Adam, Adamax, and Nadam. See https://keras.io/optimizers/"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"model.compile(loss='categorical_crossentropy',\n",
" optimizer=RMSprop(),\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Training our model is also just one line of code with Keras. Here we'll do 10 epochs with a batch size of 100. Keras is slower, and if we're not running on top of a GPU-accelerated Tensorflow this can take a fair amount of time (that's why I've limited it to just 10 epochs.)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9290 - loss: 0.2449 - val_accuracy: 0.9644 - val_loss: 0.1190\n",
"Epoch 2/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9700 - loss: 0.0999 - val_accuracy: 0.9731 - val_loss: 0.0854\n",
"Epoch 3/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9800 - loss: 0.0665 - val_accuracy: 0.9777 - val_loss: 0.0737\n",
"Epoch 4/10\n",
"600/600 - 1s - 2ms/step - accuracy: 0.9849 - loss: 0.0484 - val_accuracy: 0.9783 - val_loss: 0.0708\n",
"Epoch 5/10\n",
"600/600 - 1s - 2ms/step - accuracy: 0.9891 - loss: 0.0354 - val_accuracy: 0.9807 - val_loss: 0.0639\n",
"Epoch 6/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9920 - loss: 0.0267 - val_accuracy: 0.9803 - val_loss: 0.0668\n",
"Epoch 7/10\n",
"600/600 - 1s - 2ms/step - accuracy: 0.9940 - loss: 0.0201 - val_accuracy: 0.9821 - val_loss: 0.0579\n",
"Epoch 8/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9955 - loss: 0.0156 - val_accuracy: 0.9813 - val_loss: 0.0630\n",
"Epoch 9/10\n",
"600/600 - 2s - 3ms/step - accuracy: 0.9968 - loss: 0.0117 - val_accuracy: 0.9815 - val_loss: 0.0713\n",
"Epoch 10/10\n",
"600/600 - 1s - 2ms/step - accuracy: 0.9977 - loss: 0.0089 - val_accuracy: 0.9812 - val_loss: 0.0662\n"
]
}
],
"source": [
"history = model.fit(train_images, train_labels,\n",
" batch_size=100,\n",
" epochs=10,\n",
" verbose=2,\n",
" validation_data=(test_images, test_labels))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"But, even with just 10 epochs, we've outperformed our Tensorflow version considerably!"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Test loss: 0.06624007225036621\n",
"Test accuracy: 0.9811999797821045\n"
]
}
],
"source": [
"score = model.evaluate(test_images, test_labels, verbose=0)\n",
"print('Test loss:', score[0])\n",
"print('Test accuracy:', score[1])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As before let's visualize the ones it got wrong. As this model is much better, we'll have to search deeper to find mistakes to look at."
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 42ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAaEAAAGxCAYAAADLfglZAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjkuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8hTgPZAAAACXBIWXMAAA9hAAAPYQGoP6dpAAAl50lEQVR4nO3df3RU5Z3H8c9AkgmQSWyEZBIIaaSwUqF0AQWyQhKFYFyzQHBXarcb2pUDCriILFtKPQS1BFFYeoroWbtFEVDcBVkRFVIhARuwkYMrooeGGiQCaSRKJkFICHn2D05mHRJ+3DDJkx/v1zn3HOfO/d77nYfrfHLn3rnjMsYYAQBgQRfbDQAAOi9CCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCJf14osvyuVy+aeQkBD16dNHP/3pT3X8+PFW6eG73/2upk6d6n+cn58vl8ul/Px8R+spLCxUTk6OTp8+3ei51NRUpaamXlefwfbHP/5R48ePl8fjUUREhNLS0vSHP/zhmmpzcnLkcrl06tSp6+7j6NGjcrlceuaZZ657XZeu88UXXwzK+n7729/K5XIpIiIiKOtD6yKEcFVr1qzR3r17lZeXp2nTpumVV17R6NGjdebMmVbvZejQodq7d6+GDh3qqK6wsFCLFy9uMoRWr16t1atXB6nD61dUVKQxY8bo7Nmzevnll/Xyyy/r3LlzuvPOO7V3717b7bUpx48f17x58xQfH2+7FTRTiO0G0PYNGjRIw4cPlySlpaXpwoULeuKJJ7Rlyxb9+Mc/brLmm2++Uffu3YPeS2RkpEaOHBnUdX7/+98P6vqu12OPPaYbbrhB77zzjn8Mx44dq5tuuknz5s275iOizmDGjBkaM2aMoqOj9d///d+220EzcCQExxpC4PPPP5ckTZ06VRERETp48KDS09Pl8Xh05513SpJqa2v15JNP6uabb5bb7VavXr3005/+VF9++WXAOs+fP6/58+fL6/Wqe/fuuv322/XHP/6x0bYv93Hc+++/r8zMTN14440KDw9Xv379NGfOHEkXP57613/9V0lSUlKS/+PFhnU09XHcV199pYceeki9e/dWWFiYbrrpJi1cuFA1NTUBy7lcLs2aNUsvv/yyBg4cqO7du2vIkCF68803HY9rgz/84Q9KTU0NCHGPx6MxY8aosLBQJ0+ebPa6G3z55Zd66KGH9P3vf18RERGKiYnRHXfcoT179jS5fH19vX71q1+pb9++Cg8P1/Dhw/Xuu+82Wq64uFj333+/YmJi5Ha7NXDgQD377LPX3W9T1q1bp4KCgjZ1FAvnOBKCY0eOHJEk9erVyz+vtrZWf/d3f6fp06fr5z//uerq6lRfX68JEyZoz549mj9/vpKTk/X5559r0aJFSk1N1QcffKBu3bpJkqZNm6a1a9dq3rx5GjdunD7++GNlZWWpqqrqqv1s375dmZmZGjhwoFasWKG+ffvq6NGj2rFjhyTpgQce0FdffaXf/OY32rx5s+Li4iRd/gjo3LlzSktL05///GctXrxYP/jBD7Rnzx7l5ubqww8/1LZt2wKW37Ztm4qKivT4448rIiJCy5Yt06RJk3T48GHddNNN/uVcLpdSUlKuej6rtrZWbre70fyGeQcPHvS/hub66quvJEmLFi2S1+tVdXW1Xn/9daWmpurdd99tFMqrVq1SYmKiVq5cqfr6ei1btkwZGRkqKCjQqFGjJEmffPKJkpOT1bdvXy1fvlxer1fbt2/Xww8/rFOnTmnRokVX7Olax0eSysvLNWfOHC1dulR9+vRp1higjTDAZaxZs8ZIMvv27TPnz583VVVV5s033zS9evUyHo/HlJWVGWOMyc7ONpLM7373u4D6V155xUgymzZtCphfVFRkJJnVq1cbY4z59NNPjSTzyCOPBCy3fv16I8lkZ2f75+3atctIMrt27fLP69evn+nXr585e/bsZV/L008/bSSZkpKSRs+lpKSYlJQU/+Pnn3/eSDKvvfZawHJPPfWUkWR27NjhnyfJxMbGGp/P559XVlZmunTpYnJzcwPqu3btau64447L9tjghz/8oRkwYIC5cOGCf9758+fNTTfdZCSZDRs2XLF+0aJFRpL58ssvr7qtBnV1deb8+fPmzjvvNJMmTfLPLykpMZJMfHx8wPj6fD4THR1txo4d6583fvx406dPH1NZWRmw7lmzZpnw8HDz1VdfBaxzzZo1Actd6/gYY8zkyZNNcnKyqa+vN8Zc3Ad79Ohxza8XbQcfx+GqRo4cqdDQUHk8Ht1zzz3yer16++23FRsbG7Dc5MmTAx6/+eabuuGGG5SZmam6ujr/9MMf/lBer9f/F++uXbskqdH5pX/4h39QSMiVD9b/9Kc/6c9//rP++Z//WeHh4df5Si/auXOnevTooXvvvTdgfsNVepd+DJWWliaPx+N/HBsbq5iYGP/HlQ3q6uqa/AjrUrNnz9af/vQnzZo1S8ePH1dpaalmzJjhX1+XLsH53/b555/X0KFDFR4erpCQEIWGhurdd9/Vp59+2mjZrKysgPH1eDzKzMzU7t27deHCBZ07d07vvvuuJk2apO7duwf8e9999906d+6c9u3bd8V+rnV8Nm3apK1bt+qFF16Qy+Vy/sLRphBCuKq1a9eqqKhIBw4c0IkTJ/TRRx/pb/7mbwKW6d69uyIjIwPm/eUvf9Hp06cVFham0NDQgKmsrMx/CXFFRYUkyev1BtSHhIToxhtvvGJvDeeWgvmRTEVFhbxeb6M3uJiYGIWEhPj7bdBUj263W2fPnm3W9n/2s59p6dKlevnll9WnTx/17dtXn3zyiebNmydJ6t27d7PW+20rVqzQgw8+qBEjRmjTpk3at2+fioqKdNdddzXZ96X/Ng3zamtrVV1drYqKCtXV1ek3v/lNo3/ru+++W5KCcsl4dXW1Zs6cqdmzZys+Pl6nT5/W6dOnVVtbK0k6ffq0las20XycE8JVDRw40H913OU09Rdpz549deONN+qdd95psqbh6KHhTbysrCzgDbaurq7RG/6lGs5LffHFF1dczokbb7xR77//vowxAa+rvLxcdXV16tmzZ9C2dTn/9m//pjlz5qi4uFgej0eJiYmaPn26evTooWHDhl33+tetW6fU1FQ999xzAfMvdw6urKysyXlhYWGKiIhQaGiounbtqp/85CeaOXNmk+tISkq67r5PnTqlv/zlL1q+fLmWL1/e6PnvfOc7mjBhgrZs2XLd20LrIITQYu655x69+uqrunDhgkaMGHHZ5RpOgq9fvz7gDfa1115TXV3dFbcxYMAA9evXT7/73e80d+7cJk/oS/9/Uv9ajk7uvPNOvfbaa9qyZYsmTZrkn7927Vr/863B7XZr0KBBkqRjx45p48aNmjZtmv9ijuvhcrkajdVHH32kvXv3KiEhodHymzdv1tNPP+3/SK6qqkpbt27V6NGj1bVrV3Xv3l1paWk6cOCAfvCDHygsLOy6e2yK1+v1f3z7bUuXLlVBQYHefvvtVvkjAcFDCKHFTJkyRevXr9fdd9+tf/mXf9Ftt92m0NBQffHFF9q1a5cmTJigSZMmaeDAgfrHf/xHrVy5UqGhoRo7dqw+/vhjPfPMM40+4mvKs88+q8zMTI0cOVKPPPKI+vbtq2PHjmn79u1av369JGnw4MGSpF//+tfKzs5WaGio/uqv/irgXE6Df/qnf9Kzzz6r7OxsHT16VIMHD9Z7772nJUuW6O6779bYsWObNR4hISFKSUm56nmPjz/+WJs2bdLw4cPldrv1v//7v1q6dKn69++vJ5544pq3t3Xr1iZf37333qt77rlHTzzxhBYtWqSUlBQdPnxYjz/+uJKSkpoM/q5du2rcuHGaO3eu6uvr9dRTT8nn82nx4sX+ZX7961/r9ttv1+jRo/Xggw/qu9/9rqqqqnTkyBFt3bpVO3fuvGK/1zI+4eHhTd7d4sUXX1TXrl3b3J0vcA1sXxmBtqvh6riioqIrLnelK5POnz9vnnnmGTNkyBATHh5uIiIizM0332ymT59uiouL/cvV1NSYRx991MTExJjw8HAzcuRIs3fvXpOYmHjVq+OMMWbv3r0mIyPDREVFGbfbbfr169foarsFCxaY+Ph406VLl4B1XHp1nDHGVFRUmBkzZpi4uDgTEhJiEhMTzYIFC8y5c+cClpNkZs6c2eh1X9p3w7KXbqcphw8fNmPGjDHR0dEmLCzMfO973zO//OUvTXV19VVrjfn/q+MuNxlzcbznzZtnevfubcLDw83QoUPNli1bTHZ2tklMTPSvq+FKtqeeesosXrzY9OnTx4SFhZm//uu/Ntu3b2+07ZKSEvOzn/3M9O7d24SGhppevXqZ5ORk8+STTzZa56VXx13r+DSFq+PaL5cxxljIPgAAuDoOAGAPIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABr2tyXVevr63XixAl5PB5uTggA7ZAxRlVVVYqPj7/qDXfbXAidOHGiyduGAADal9LS0qveXLjNhVDDbUZKS0uv6ZYtAIC2xefzKSEhocnbRl2qxUJo9erVevrpp3Xy5EndcsstWrlypUaPHn3VuoaP4CIjIwkhAGjHruWUSotcmLBx40bNmTNHCxcu1IEDBzR69GhlZGTo2LFjLbE5AEA71SL3jhsxYoSGDh0a8FslAwcO1MSJE5Wbm3vFWp/Pp6ioKFVWVnIkBADtkJP38aAfCdXW1mr//v1KT08PmJ+enq7CwsJGy9fU1Mjn8wVMAIDOIeghdOrUKV24cEGxsbEB82NjY5v8dcbc3FxFRUX5J66MA4DOo8W+rHrpCSlzyU8lN1iwYIEqKyv9U2lpaUu1BABoY4J+dVzPnj3VtWvXRkc95eXljY6OpIs/YXy5n2QGAHRsQT8SCgsL07Bhw5SXlxcwPy8vT8nJycHeHACgHWuR7wnNnTtXP/nJTzR8+HCNGjVK//Ef/6Fjx45pxowZLbE5AEA71SIhdN9996miokKPP/64Tp48qUGDBumtt95SYmJiS2wOANBOtcj3hK4H3xMCgPbN6veEAAC4VoQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWhNhuAMC1ef755x3XPPjgg83a1qZNmxzXZGVlNWtb6Nw4EgIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAa7iBKWDB6tWrHdfMmjWrBTppmsfjabVtoXPjSAgAYA0hBACwJughlJOTI5fLFTB5vd5gbwYA0AG0yDmhW265Rb///e/9j7t27doSmwEAtHMtEkIhISEc/QAArqpFzgkVFxcrPj5eSUlJmjJlij777LPLLltTUyOfzxcwAQA6h6CH0IgRI7R27Vpt375dL7zwgsrKypScnKyKiooml8/NzVVUVJR/SkhICHZLAIA2KughlJGRocmTJ2vw4MEaO3astm3bJkl66aWXmlx+wYIFqqys9E+lpaXBbgkA0Ea1+JdVe/ToocGDB6u4uLjJ591ut9xud0u3AQBog1r8e0I1NTX69NNPFRcX19KbAgC0M0EPoXnz5qmgoEAlJSV6//33de+998rn8yk7OzvYmwIAtHNB/zjuiy++0I9+9COdOnVKvXr10siRI7Vv3z4lJiYGe1MAgHbOZYwxtpv4Np/Pp6ioKFVWVioyMtJ2O8BVFRYWOq5JSUlxXBMS4vxvxnXr1jmukaTJkyc3qw6QnL2Pc+84AIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCmxX/UDmhPPvnkE8c1U6ZMaYFOGlu2bJnjGm5EiraOIyEAgDWEEADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYw1200SEdPXq0WXXjx493XHPixAnHNf/+7//uuGb27NmOa4C2jiMhAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGG5iizaurq3NcM3fu3GZt6/jx445rHnnkEcc1Dz/8sOOa1nThwgXHNV26OP+b1uVyOa5Bx8KREADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYww1M0eatXLnScc3rr7/erG1NmTLFcc3y5cubta3WUF9f36y65ozD+PHjHdc88MADjmvQsXAkBACwhhACAFjjOIR2796tzMxMxcfHy+VyacuWLQHPG2OUk5Oj+Ph4devWTampqTp06FCw+gUAdCCOQ+jMmTMaMmSIVq1a1eTzy5Yt04oVK7Rq1SoVFRXJ6/Vq3Lhxqqqquu5mAQAdi+MLEzIyMpSRkdHkc8YYrVy5UgsXLlRWVpYk6aWXXlJsbKw2bNig6dOnX1+3AIAOJajnhEpKSlRWVqb09HT/PLfbrZSUFBUWFjZZU1NTI5/PFzABADqHoIZQWVmZJCk2NjZgfmxsrP+5S+Xm5ioqKso/JSQkBLMlAEAb1iJXx7lcroDHxphG8xosWLBAlZWV/qm0tLQlWgIAtEFB/bKq1+uVdPGIKC4uzj+/vLy80dFRA7fbLbfbHcw2AADtRFCPhJKSkuT1epWXl+efV1tbq4KCAiUnJwdzUwCADsDxkVB1dbWOHDnif1xSUqIPP/xQ0dHR6tu3r+bMmaMlS5aof//+6t+/v5YsWaLu3bvr/vvvD2rjAID2z3EIffDBB0pLS/M/njt3riQpOztbL774oubPn6+zZ8/qoYce0tdff60RI0Zox44d8ng8wesaANAhuIwxxnYT3+bz+RQVFaXKykpFRkbabgdB9vnnnzuuGT16tOOaG264wXGNJL322muOa26++eZmbas1HDt2rFl1iYmJjmtuueUWxzUffPCB45rw8HDHNWhdTt7HuXccAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArAnqL6sCV7N06VLHNc35yfdHH33UcY3Utu+Iff78ecc1CxcubIFOmna5X0++Eu6IDY6EAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAabmCKZjty5IjjmjVr1jiu+du//VvHNQ8//LDjmrauOTdyXbduXQt00rSJEye22rbQcXAkBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWcANTNNvWrVsd19TU1Diuqa+vd1zTEf3Xf/2X7Rau6O///u9tt4B2iCMhAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGG5ii2caPH++4JiTE+S739ttvO66ZPHmy4xpJ+sUvfuG4Zvjw4Y5riouLHdc8+eSTjmua64EHHnBcExMT0wKdoKPjSAgAYA0hBACwxnEI7d69W5mZmYqPj5fL5dKWLVsCnp86dapcLlfANHLkyGD1CwDoQByH0JkzZzRkyBCtWrXqssvcddddOnnypH966623rqtJAEDH5PgscUZGhjIyMq64jNvtltfrbXZTAIDOoUXOCeXn5ysmJkYDBgzQtGnTVF5eftlla2pq5PP5AiYAQOcQ9BDKyMjQ+vXrtXPnTi1fvlxFRUW64447VFNT0+Tyubm5ioqK8k8JCQnBbgkA0EYF/XtC9913n/+/Bw0apOHDhysxMVHbtm1TVlZWo+UXLFiguXPn+h/7fD6CCAA6iRb/smpcXJwSExMv++U8t9stt9vd0m0AANqgFv+eUEVFhUpLSxUXF9fSmwIAtDOOj4Sqq6t15MgR/+OSkhJ9+OGHio6OVnR0tHJycjR58mTFxcXp6NGj+sUvfqGePXtq0qRJQW0cAND+OQ6hDz74QGlpaf7HDedzsrOz9dxzz+ngwYNau3atTp8+rbi4OKWlpWnjxo3yeDzB6xoA0CG4jDHGdhPf5vP5FBUVpcrKSkVGRtpuB0HWnJtwPvbYYy3QSdNuuOEGxzW33Xab45rCwkLHNdXV1Y5runRp3ifux48fd1zDdwPRwMn7OPeOAwBYQwgBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDUt/suqwLf98pe/dFzzve99z3HN/PnzHddIUmlpqeOaHTt2NGtbrWHMmDHNquOO2GgtHAkBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDUuY4yx3cS3+Xw+RUVFqbKyUpGRkbbbQTtVW1vbrLoLFy44rjl06JDjmltvvdVxTY8ePRzXNKc3SUpMTGxWHSA5ex/nSAgAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArAmx3QDQEsLCwlptW9XV1a2ynbi4OMc13IgUbR1HQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDTcwBa7TsmXLWmU748ePb5XtAK2JIyEAgDWEEADAGkchlJubq1tvvVUej0cxMTGaOHGiDh8+HLCMMUY5OTmKj49Xt27dlJqaqkOHDgW1aQBAx+AohAoKCjRz5kzt27dPeXl5qqurU3p6us6cOeNfZtmyZVqxYoVWrVqloqIieb1ejRs3TlVVVUFvHgDQvjm6MOGdd94JeLxmzRrFxMRo//79GjNmjIwxWrlypRYuXKisrCxJ0ksvvaTY2Fht2LBB06dPD17nAIB277rOCVVWVkqSoqOjJUklJSUqKytTenq6fxm3262UlBQVFhY2uY6amhr5fL6ACQDQOTQ7hIwxmjt3rm6//XYNGjRIklRWViZJio2NDVg2NjbW/9ylcnNzFRUV5Z8SEhKa2xIAoJ1pdgjNmjVLH330kV555ZVGz7lcroDHxphG8xosWLBAlZWV/qm0tLS5LQEA2plmfVl19uzZeuONN7R792716dPHP9/r9Uq6eEQUFxfnn19eXt7o6KiB2+2W2+1uThsAgHbO0ZGQMUazZs3S5s2btXPnTiUlJQU8n5SUJK/Xq7y8PP+82tpaFRQUKDk5OTgdAwA6DEdHQjNnztSGDRv0P//zP/J4PP7zPFFRUerWrZtcLpfmzJmjJUuWqH///urfv7+WLFmi7t276/7772+RFwAAaL8chdBzzz0nSUpNTQ2Yv2bNGk2dOlWSNH/+fJ09e1YPPfSQvv76a40YMUI7duyQx+MJSsMAgI7DUQgZY666jMvlUk5OjnJycprbE2DNhx9+6Ljm97//ffAbAToJ7h0HALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAa5r1y6pAR1VdXe245vz5845rIiIiHNf8+Mc/dlwDtHUcCQEArCGEAADWEEIAAGsIIQCANYQQAMAaQggAYA0hBACwhhACAFhDCAEArCGEAADWEEIAAGsIIQCANdzAFPiWmJgYxzXdunVzXDNs2DDHNaNGjXJcA7R1HAkBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDXcwBT4lgEDBjiu+eabb1qgE6Bz4EgIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDWEEADAGkIIAGANIQQAsIYQAgBYQwgBAKwhhAAA1hBCAABrCCEAgDWOQig3N1e33nqrPB6PYmJiNHHiRB0+fDhgmalTp8rlcgVMI0eODGrTAICOwVEIFRQUaObMmdq3b5/y8vJUV1en9PR0nTlzJmC5u+66SydPnvRPb731VlCbBgB0DI5+WfWdd94JeLxmzRrFxMRo//79GjNmjH++2+2W1+sNTocAgA7rus4JVVZWSpKio6MD5ufn5ysmJkYDBgzQtGnTVF5eftl11NTUyOfzBUwAgM7BZYwxzSk0xmjChAn6+uuvtWfPHv/8jRs3KiIiQomJiSopKdFjjz2muro67d+/X263u9F6cnJytHjx4kbzKysrFRkZ2ZzWAAAW+Xw+RUVFXdP7eLNDaObMmdq2bZvee+899enT57LLnTx5UomJiXr11VeVlZXV6PmamhrV1NQENJ+QkEAIAUA75SSEHJ0TajB79my98cYb2r179xUDSJLi4uKUmJio4uLiJp93u91NHiEBADo+RyFkjNHs2bP1+uuvKz8/X0lJSVetqaioUGlpqeLi4prdJACgY3J0YcLMmTO1bt06bdiwQR6PR2VlZSorK9PZs2clSdXV1Zo3b5727t2ro0ePKj8/X5mZmerZs6cmTZrUIi8AANB+OTon5HK5mpy/Zs0aTZ06VWfPntXEiRN14MABnT59WnFxcUpLS9MTTzyhhISEa9qGk88SAQBtT4udE7paXnXr1k3bt293skoAQCfGveMAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCsIYQAANaE2G7gUsYYSZLP57PcCQCgORrevxvez6+kzYVQVVWVJCkhIcFyJwCA61FVVaWoqKgrLuMy1xJVrai+vl4nTpyQx+ORy+UKeM7n8ykhIUGlpaWKjIy01KF9jMNFjMNFjMNFjMNFbWEcjDGqqqpSfHy8unS58lmfNnck1KVLF/Xp0+eKy0RGRnbqnawB43AR43AR43AR43CR7XG42hFQAy5MAABYQwgBAKxpVyHkdru1aNEiud1u261YxThcxDhcxDhcxDhc1N7Goc1dmAAA6Dza1ZEQAKBjIYQAANYQQgAAawghAIA1hBAAwJp2FUKrV69WUlKSwsPDNWzYMO3Zs8d2S60qJydHLpcrYPJ6vbbbanG7d+9WZmam4uPj5XK5tGXLloDnjTHKyclRfHy8unXrptTUVB06dMhOsy3oauMwderURvvHyJEj7TTbQnJzc3XrrbfK4/EoJiZGEydO1OHDhwOW6Qz7w7WMQ3vZH9pNCG3cuFFz5szRwoULdeDAAY0ePVoZGRk6duyY7dZa1S233KKTJ0/6p4MHD9puqcWdOXNGQ4YM0apVq5p8ftmyZVqxYoVWrVqloqIieb1ejRs3zn8z3I7iauMgSXfddVfA/vHWW2+1Yoctr6CgQDNnztS+ffuUl5enuro6paen68yZM/5lOsP+cC3jILWT/cG0E7fddpuZMWNGwLybb77Z/PznP7fUUetbtGiRGTJkiO02rJJkXn/9df/j+vp64/V6zdKlS/3zzp07Z6Kioszzzz9vocPWcek4GGNMdna2mTBhgpV+bCkvLzeSTEFBgTGm8+4Pl46DMe1nf2gXR0K1tbXav3+/0tPTA+anp6ersLDQUld2FBcXKz4+XklJSZoyZYo+++wz2y1ZVVJSorKysoB9w+12KyUlpdPtG5KUn5+vmJgYDRgwQNOmTVN5ebntllpUZWWlJCk6OlpS590fLh2HBu1hf2gXIXTq1ClduHBBsbGxAfNjY2NVVlZmqavWN2LECK1du1bbt2/XCy+8oLKyMiUnJ6uiosJ2a9Y0/Pt39n1DkjIyMrR+/Xrt3LlTy5cvV1FRke644w7V1NTYbq1FGGM0d+5c3X777Ro0aJCkzrk/NDUOUvvZH9rcTzlcyaW/L2SMaTSvI8vIyPD/9+DBgzVq1Cj169dPL730kubOnWuxM/s6+74hSffdd5//vwcNGqThw4crMTFR27ZtU1ZWlsXOWsasWbP00Ucf6b333mv0XGfaHy43Du1lf2gXR0I9e/ZU165dG/0lU15e3ugvns6kR48eGjx4sIqLi223Yk3D1YHsG43FxcUpMTGxQ+4fs2fP1htvvKFdu3YF/P5YZ9sfLjcOTWmr+0O7CKGwsDANGzZMeXl5AfPz8vKUnJxsqSv7ampq9OmnnyouLs52K9YkJSXJ6/UG7Bu1tbUqKCjo1PuGJFVUVKi0tLRD7R/GGM2aNUubN2/Wzp07lZSUFPB8Z9kfrjYOTWmz+4PFiyIcefXVV01oaKj5z//8T/PJJ5+YOXPmmB49epijR4/abq3VPProoyY/P9989tlnZt++feaee+4xHo+nw49BVVWVOXDggDlw4ICRZFasWGEOHDhgPv/8c2OMMUuXLjVRUVFm8+bN5uDBg+ZHP/qRiYuLMz6fz3LnwXWlcaiqqjKPPvqoKSwsNCUlJWbXrl1m1KhRpnfv3h1qHB588EETFRVl8vPzzcmTJ/3TN99841+mM+wPVxuH9rQ/tJsQMsaYZ5991iQmJpqwsDAzdOjQgMsRO4P77rvPxMXFmdDQUBMfH2+ysrLMoUOHbLfV4nbt2mUkNZqys7ONMRcvy120aJHxer3G7XabMWPGmIMHD9ptugVcaRy++eYbk56ebnr16mVCQ0NN3759TXZ2tjl27JjttoOqqdcvyaxZs8a/TGfYH642Du1pf+D3hAAA1rSLc0IAgI6JEAIAWEMIAQCsIYQAANYQQgAAawghAIA1hBAAwBpCCABgDSEEALCGEAIAWEMIAQCs+T9JzsrvEeOeJQAAAABJRU5ErkJggg==",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 53ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 46ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 23ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 61ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 27ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n"
]
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 640x480 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 24ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 26ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 25ms/step\n",
"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 28ms/step\n"
]
}
],
"source": [
"for x in range(1000):\n",
" test_image = test_images[x,:].reshape(1,784)\n",
" predicted_cat = model.predict(test_image).argmax()\n",
" label = test_labels[x].argmax()\n",
" if (predicted_cat != label):\n",
" plt.title('Prediction: %d Label: %d' % (predicted_cat, label))\n",
" plt.imshow(test_image.reshape([28,28]), cmap=plt.get_cmap('gray_r'))\n",
" plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true,
"jupyter": {
"outputs_hidden": true
}
},
"source": [
"You can see most of the ones it's having trouble with, are images a human would have trouble with as well!\n",
"\n",
"## Excercise\n",
"\n",
"As before, see if you can improve on the results! Does running more epochs help considerably? How about trying different optimizers?\n",
"\n",
"You can also take advantage of Keras's ease of use to try different topologies quickly. Keras includes a MNIST example, where they add an additional layer, and use Dropout at each step to prevent overfitting, like this:\n",
"\n",
"`\n",
"model = Sequential()\n",
"model.add(Dense(512, activation='relu', input_shape=(784,)))\n",
"model.add(Dropout(0.2))\n",
"model.add(Dense(512, activation='relu'))\n",
"model.add(Dropout(0.2))\n",
"model.add(Dense(10, activation='softmax'))\n",
"`\n",
"\n",
"Try adapting that to our code above and see if it makes a difference or not."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python [conda env:base] *",
"language": "python",
"name": "conda-base-py"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.12.7"
},
"widgets": {
"state": {},
"version": "1.1.2"
}
},
"nbformat": 4,
"nbformat_minor": 4
}