200 lines
17 KiB
Plaintext
200 lines
17 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 27,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import numpy as np\n",
|
|
"import tensorflow as tf\n",
|
|
"import keras\n",
|
|
"import random\n",
|
|
"from matplotlib import pyplot as plt\n",
|
|
"%matplotlib inline"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Exercise 9\n",
|
|
"Mess with Fashion MNIST, TFRecords, and TF dataset"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 28,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# Grab our data\n",
|
|
"fashion_mnist = keras.datasets.fashion_mnist\n",
|
|
"(X_train_full, y_train_full), (X_test, y_test) = fashion_mnist.load_data()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 29,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"(60000, 28, 28)\n",
|
|
"(60000,)\n",
|
|
"(10000, 28, 28)\n",
|
|
"(10000,)\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"print(X_train_full.shape)\n",
|
|
"print(y_train_full.shape)\n",
|
|
"print(X_test.shape)\n",
|
|
"print(y_test.shape)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 31,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"2\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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\n",
|
|
"text/plain": [
|
|
"<Figure size 432x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {
|
|
"needs_background": "light"
|
|
},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Grab random image to display\n",
|
|
"idx = random.randint(0, X_train_full.shape[0])\n",
|
|
"plt.imshow(X_train_full[idx])\n",
|
|
"print(y_train_full[idx])"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 32,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# Split training set into training and validation sets and normalize pixel range\n",
|
|
"X_valid, X_train = X_train_full[:5000] / 255., X_train_full[5000:] / 255.\n",
|
|
"y_valid, y_train = y_train_full[:5000], y_train_full[5000:]\n",
|
|
"X_test = X_test / 255."
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 33,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"class_names = [\"T-shirt/top\", \"Trouser\", \"Pullover\", \"Dress\", \"Coat\",\n",
|
|
" \"Sandal\", \"Shirt\", \"Sneaker\", \"Bag\", \"Ankle boot\"]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 34,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"# Shuffle training and validation sets\n",
|
|
"from sklearn.utils import shuffle\n",
|
|
"X_valid, y_valid = shuffle(X_valid, y_valid)\n",
|
|
"X_train, y_train = shuffle(X_train, y_train)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 35,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"8\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"image/png": "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\n",
|
|
"text/plain": [
|
|
"<Figure size 432x288 with 1 Axes>"
|
|
]
|
|
},
|
|
"metadata": {
|
|
"needs_background": "light"
|
|
},
|
|
"output_type": "display_data"
|
|
}
|
|
],
|
|
"source": [
|
|
"# Store using TFRecord \n",
|
|
"sets = [[x_valid, 'x_valid'], [x_train, 'x_train'], [y_valid, 'y_valid'], [y_train, 'y_train']]\n",
|
|
"for set_ in sets:\n",
|
|
" data = set_[0]\n",
|
|
" filename = set_[1]\n",
|
|
" # TODO go through each set and store as TFRecord\n",
|
|
" \n",
|
|
"\n",
|
|
" "
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Exercise 10\n",
|
|
"Download a dataset, split it up, create a tf.data.dataset out of it, then build binary classification model using an Embedding Layer"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3",
|
|
"language": "python",
|
|
"name": "python3"
|
|
},
|
|
"language_info": {
|
|
"codemirror_mode": {
|
|
"name": "ipython",
|
|
"version": 3
|
|
},
|
|
"file_extension": ".py",
|
|
"mimetype": "text/x-python",
|
|
"name": "python",
|
|
"nbconvert_exporter": "python",
|
|
"pygments_lexer": "ipython3",
|
|
"version": "3.7.6"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|