Showing preview only (7,839K chars total). Download the full file or copy to clipboard to get everything.
Repository: Kulbear/deep-learning-coursera
Branch: master
Commit: 997fdb2e2db6
Files: 33
Total size: 28.3 MB
Directory structure:
gitextract_stf4uzcu/
├── .gitignore
├── Convolutional Neural Networks/
│ ├── Convolution model - Application - v1.ipynb
│ ├── Convolution model - Step by Step - v1.ipynb
│ ├── Keras - Tutorial - Happy House v1.ipynb
│ └── Residual Networks - v1.ipynb
├── Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/
│ ├── Gradient Checking.ipynb
│ ├── Initialization.ipynb
│ ├── Optimization methods.ipynb
│ ├── Regularization.ipynb
│ ├── Tensorflow Tutorial.ipynb
│ ├── Week 1 Quiz - Practical aspects of deep learning.md
│ ├── Week 2 Quiz - Optimization algorithms.md
│ └── Week 3 Quiz - Hyperparameter tuning, Batch Normalization, Programming Frameworks.md
├── LICENSE
├── Neural Networks and Deep Learning/
│ ├── Building your Deep Neural Network - Step by Step.ipynb
│ ├── Deep Neural Network - Application.ipynb
│ ├── Logistic Regression with a Neural Network mindset.ipynb
│ ├── Planar data classification with one hidden layer.ipynb
│ ├── Week 1 Quiz - Introduction to deep learning.md
│ ├── Week 2 Quiz - Neural Network Basics.md
│ ├── Week 3 Quiz - Shallow Neural Networks.md
│ └── Week 4 Quiz - Key concepts on Deep Neural Networks.md
├── README.md
├── Sequence Models/
│ ├── Building a Recurrent Neural Network - Step by Step - v2.ipynb
│ ├── Dinosaurus Island -- Character level language model final - v3.ipynb
│ ├── Emojify - v2.ipynb
│ ├── Improvise a Jazz Solo with an LSTM Network - v1.ipynb
│ ├── Neural machine translation with attention - v2.ipynb
│ ├── Operations on word vectors - v2.ipynb
│ ├── Trigger word detection - v1.ipynb
│ └── rnn_utils.py
└── Structuring Machine Learning Projects/
├── Week 1 Quiz - Bird recognition in the city of Peacetopia (case study).md
└── Week 2 Quiz - Autonomous driving (case study).md
================================================
FILE CONTENTS
================================================
================================================
FILE: .gitignore
================================================
# Byte-compiled / optimized / DLL files
__pycache__/
*.py[cod]
*$py.class
# C extensions
*.so
.idea
.DS_Store
# Distribution / packaging
.Python
env/
build/
develop-eggs/
dist/
downloads/
eggs/
.eggs/
lib/
lib64/
parts/
sdist/
var/
wheels/
*.egg-info/
.installed.cfg
*.egg
# PyInstaller
# Usually these files are written by a python script from a template
# before PyInstaller builds the exe, so as to inject date/other infos into it.
*.manifest
*.spec
# Installer logs
pip-log.txt
pip-delete-this-directory.txt
# Unit test / coverage reports
htmlcov/
.tox/
.coverage
.coverage.*
.cache
nosetests.xml
coverage.xml
*.cover
.hypothesis/
# Translations
*.mo
*.pot
# Django stuff:
*.log
local_settings.py
# Flask stuff:
instance/
.webassets-cache
# Scrapy stuff:
.scrapy
# Sphinx documentation
docs/_build/
# PyBuilder
target/
# Jupyter Notebook
.ipynb_checkpoints
# pyenv
.python-version
# celery beat schedule file
celerybeat-schedule
# SageMath parsed files
*.sage.py
# dotenv
.env
# virtualenv
.venv
venv/
ENV/
# Spyder project settings
.spyderproject
.spyproject
# Rope project settings
.ropeproject
# mkdocs documentation
/site
# mypy
.mypy_cache/
================================================
FILE: Convolutional Neural Networks/Convolution model - Application - v1.ipynb
================================================
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Convolutional Neural Networks: Application\n",
"\n",
"Welcome to Course 4's second assignment! In this notebook, you will:\n",
"\n",
"- Implement helper functions that you will use when implementing a TensorFlow model\n",
"- Implement a fully functioning ConvNet using TensorFlow \n",
"\n",
"**After this assignment you will be able to:**\n",
"\n",
"- Build and train a ConvNet in TensorFlow for a classification problem \n",
"\n",
"We assume here that you are already familiar with TensorFlow. If you are not, please refer the *TensorFlow Tutorial* of the third week of Course 2 (\"*Improving deep neural networks*\")."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1.0 - TensorFlow model\n",
"\n",
"In the previous assignment, you built helper functions using numpy to understand the mechanics behind convolutional neural networks. Most practical applications of deep learning today are built using programming frameworks, which have many built-in functions you can simply call. \n",
"\n",
"As usual, we will start by loading in the packages. "
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import math\n",
"import numpy as np\n",
"import h5py\n",
"import matplotlib.pyplot as plt\n",
"import scipy\n",
"from PIL import Image\n",
"from scipy import ndimage\n",
"import tensorflow as tf\n",
"from tensorflow.python.framework import ops\n",
"from cnn_utils import *\n",
"\n",
"%matplotlib inline\n",
"np.random.seed(1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Run the next cell to load the \"SIGNS\" dataset you are going to use."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# Loading the data (signs)\n",
"X_train_orig, Y_train_orig, X_test_orig, Y_test_orig, classes = load_dataset()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"As a reminder, the SIGNS dataset is a collection of 6 signs representing numbers from 0 to 5.\n",
"\n",
"<img src=\"images/SIGNS.png\" style=\"width:800px;height:300px;\">\n",
"\n",
"The next cell will show you an example of a labelled image in the dataset. Feel free to change the value of `index` below and re-run to see different examples. "
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"y = 2\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyuoTDQAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7fa8b6ac1470>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Example of a picture\n",
"index = 6\n",
"plt.imshow(X_train_orig[index])\n",
"print (\"y = \" + str(np.squeeze(Y_train_orig[:, index])))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In Course 2, you had built a fully-connected network for this dataset. But since this is an image dataset, it is more natural to apply a ConvNet to it.\n",
"\n",
"To get started, let's examine the shapes of your data. "
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"number of training examples = 1080\n",
"number of test examples = 120\n",
"X_train shape: (1080, 64, 64, 3)\n",
"Y_train shape: (1080, 6)\n",
"X_test shape: (120, 64, 64, 3)\n",
"Y_test shape: (120, 6)\n"
]
}
],
"source": [
"X_train = X_train_orig/255.\n",
"X_test = X_test_orig/255.\n",
"Y_train = convert_to_one_hot(Y_train_orig, 6).T\n",
"Y_test = convert_to_one_hot(Y_test_orig, 6).T\n",
"print (\"number of training examples = \" + str(X_train.shape[0]))\n",
"print (\"number of test examples = \" + str(X_test.shape[0]))\n",
"print (\"X_train shape: \" + str(X_train.shape))\n",
"print (\"Y_train shape: \" + str(Y_train.shape))\n",
"print (\"X_test shape: \" + str(X_test.shape))\n",
"print (\"Y_test shape: \" + str(Y_test.shape))\n",
"conv_layers = {}"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"### 1.1 - Create placeholders\n",
"\n",
"TensorFlow requires that you create placeholders for the input data that will be fed into the model when running the session.\n",
"\n",
"**Exercise**: Implement the function below to create placeholders for the input image X and the output Y. You should not define the number of training examples for the moment. To do so, you could use \"None\" as the batch size, it will give you the flexibility to choose it later. Hence X should be of dimension **[None, n_H0, n_W0, n_C0]** and Y should be of dimension **[None, n_y]**. [Hint](https://www.tensorflow.org/api_docs/python/tf/placeholder)."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: create_placeholders\n",
"\n",
"def create_placeholders(n_H0, n_W0, n_C0, n_y):\n",
" \"\"\"\n",
" Creates the placeholders for the tensorflow session.\n",
" \n",
" Arguments:\n",
" n_H0 -- scalar, height of an input image\n",
" n_W0 -- scalar, width of an input image\n",
" n_C0 -- scalar, number of channels of the input\n",
" n_y -- scalar, number of classes\n",
" \n",
" Returns:\n",
" X -- placeholder for the data input, of shape [None, n_H0, n_W0, n_C0] and dtype \"float\"\n",
" Y -- placeholder for the input labels, of shape [None, n_y] and dtype \"float\"\n",
" \"\"\"\n",
"\n",
" ### START CODE HERE ### (≈2 lines)\n",
" X = tf.placeholder(tf.float32, [None, n_H0, n_W0, n_C0])\n",
" Y = tf.placeholder(tf.float32, [None, n_y])\n",
" ### END CODE HERE ###\n",
" \n",
" return X, Y"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"X = Tensor(\"Placeholder:0\", shape=(?, 64, 64, 3), dtype=float32)\n",
"Y = Tensor(\"Placeholder_1:0\", shape=(?, 6), dtype=float32)\n"
]
}
],
"source": [
"X, Y = create_placeholders(64, 64, 3, 6)\n",
"print (\"X = \" + str(X))\n",
"print (\"Y = \" + str(Y))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**\n",
"\n",
"<table> \n",
"<tr>\n",
"<td>\n",
" X = Tensor(\"Placeholder:0\", shape=(?, 64, 64, 3), dtype=float32)\n",
"\n",
"</td>\n",
"</tr>\n",
"<tr>\n",
"<td>\n",
" Y = Tensor(\"Placeholder_1:0\", shape=(?, 6), dtype=float32)\n",
"\n",
"</td>\n",
"</tr>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1.2 - Initialize parameters\n",
"\n",
"You will initialize weights/filters $W1$ and $W2$ using `tf.contrib.layers.xavier_initializer(seed = 0)`. You don't need to worry about bias variables as you will soon see that TensorFlow functions take care of the bias. Note also that you will only initialize the weights/filters for the conv2d functions. TensorFlow initializes the layers for the fully connected part automatically. We will talk more about that later in this assignment.\n",
"\n",
"**Exercise:** Implement initialize_parameters(). The dimensions for each group of filters are provided below. Reminder - to initialize a parameter $W$ of shape [1,2,3,4] in Tensorflow, use:\n",
"```python\n",
"W = tf.get_variable(\"W\", [1,2,3,4], initializer = ...)\n",
"```\n",
"[More Info](https://www.tensorflow.org/api_docs/python/tf/get_variable)."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: initialize_parameters\n",
"\n",
"def initialize_parameters():\n",
" \"\"\"\n",
" Initializes weight parameters to build a neural network with tensorflow. The shapes are:\n",
" W1 : [4, 4, 3, 8]\n",
" W2 : [2, 2, 8, 16]\n",
" Returns:\n",
" parameters -- a dictionary of tensors containing W1, W2\n",
" \"\"\"\n",
" \n",
" tf.set_random_seed(1) # so that your \"random\" numbers match ours\n",
" \n",
" ### START CODE HERE ### (approx. 2 lines of code)\n",
" W1 = tf.get_variable(\"W1\", [4, 4, 3, 8], initializer=tf.contrib.layers.xavier_initializer(seed=0))\n",
" W2 = tf.get_variable(\"W2\", [2, 2, 8, 16], initializer=tf.contrib.layers.xavier_initializer(seed=0))\n",
" ### END CODE HERE ###\n",
"\n",
" parameters = {\"W1\": W1,\n",
" \"W2\": W2}\n",
" \n",
" return parameters"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"W1 = [ 0.00131723 0.14176141 -0.04434952 0.09197326 0.14984085 -0.03514394\n",
" -0.06847463 0.05245192]\n",
"W2 = [-0.08566415 0.17750949 0.11974221 0.16773748 -0.0830943 -0.08058\n",
" -0.00577033 -0.14643836 0.24162132 -0.05857408 -0.19055021 0.1345228\n",
" -0.22779644 -0.1601823 -0.16117483 -0.10286498]\n"
]
}
],
"source": [
"tf.reset_default_graph()\n",
"with tf.Session() as sess_test:\n",
" parameters = initialize_parameters()\n",
" init = tf.global_variables_initializer()\n",
" sess_test.run(init)\n",
" print(\"W1 = \" + str(parameters[\"W1\"].eval()[1,1,1]))\n",
" print(\"W2 = \" + str(parameters[\"W2\"].eval()[1,1,1]))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"** Expected Output:**\n",
"\n",
"<table> \n",
"\n",
" <tr>\n",
" <td>\n",
" W1 = \n",
" </td>\n",
" <td>\n",
"[ 0.00131723 0.14176141 -0.04434952 0.09197326 0.14984085 -0.03514394 <br>\n",
" -0.06847463 0.05245192]\n",
" </td>\n",
" </tr>\n",
"\n",
" <tr>\n",
" <td>\n",
" W2 = \n",
" </td>\n",
" <td>\n",
"[-0.08566415 0.17750949 0.11974221 0.16773748 -0.0830943 -0.08058 <br>\n",
" -0.00577033 -0.14643836 0.24162132 -0.05857408 -0.19055021 0.1345228 <br>\n",
" -0.22779644 -0.1601823 -0.16117483 -0.10286498]\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1.2 - Forward propagation\n",
"\n",
"In TensorFlow, there are built-in functions that carry out the convolution steps for you.\n",
"\n",
"- **tf.nn.conv2d(X,W1, strides = [1,s,s,1], padding = 'SAME'):** given an input $X$ and a group of filters $W1$, this function convolves $W1$'s filters on X. The third input ([1,f,f,1]) represents the strides for each dimension of the input (m, n_H_prev, n_W_prev, n_C_prev). You can read the full documentation [here](https://www.tensorflow.org/api_docs/python/tf/nn/conv2d)\n",
"\n",
"- **tf.nn.max_pool(A, ksize = [1,f,f,1], strides = [1,s,s,1], padding = 'SAME'):** given an input A, this function uses a window of size (f, f) and strides of size (s, s) to carry out max pooling over each window. You can read the full documentation [here](https://www.tensorflow.org/api_docs/python/tf/nn/max_pool)\n",
"\n",
"- **tf.nn.relu(Z1):** computes the elementwise ReLU of Z1 (which can be any shape). You can read the full documentation [here.](https://www.tensorflow.org/api_docs/python/tf/nn/relu)\n",
"\n",
"- **tf.contrib.layers.flatten(P)**: given an input P, this function flattens each example into a 1D vector it while maintaining the batch-size. It returns a flattened tensor with shape [batch_size, k]. You can read the full documentation [here.](https://www.tensorflow.org/api_docs/python/tf/contrib/layers/flatten)\n",
"\n",
"- **tf.contrib.layers.fully_connected(F, num_outputs):** given a the flattened input F, it returns the output computed using a fully connected layer. You can read the full documentation [here.](https://www.tensorflow.org/api_docs/python/tf/contrib/layers/fully_connected)\n",
"\n",
"In the last function above (`tf.contrib.layers.fully_connected`), the fully connected layer automatically initializes weights in the graph and keeps on training them as you train the model. Hence, you did not need to initialize those weights when initializing the parameters. \n",
"\n",
"\n",
"**Exercise**: \n",
"\n",
"Implement the `forward_propagation` function below to build the following model: `CONV2D -> RELU -> MAXPOOL -> CONV2D -> RELU -> MAXPOOL -> FLATTEN -> FULLYCONNECTED`. You should use the functions above. \n",
"\n",
"In detail, we will use the following parameters for all the steps:\n",
" - Conv2D: stride 1, padding is \"SAME\"\n",
" - ReLU\n",
" - Max pool: Use an 8 by 8 filter size and an 8 by 8 stride, padding is \"SAME\"\n",
" - Conv2D: stride 1, padding is \"SAME\"\n",
" - ReLU\n",
" - Max pool: Use a 4 by 4 filter size and a 4 by 4 stride, padding is \"SAME\"\n",
" - Flatten the previous output.\n",
" - FULLYCONNECTED (FC) layer: Apply a fully connected layer without an non-linear activation function. Do not call the softmax here. This will result in 6 neurons in the output layer, which then get passed later to a softmax. In TensorFlow, the softmax and cost function are lumped together into a single function, which you'll call in a different function when computing the cost. "
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: forward_propagation\n",
"\n",
"def forward_propagation(X, parameters):\n",
" \"\"\"\n",
" Implements the forward propagation for the model:\n",
" CONV2D -> RELU -> MAXPOOL -> CONV2D -> RELU -> MAXPOOL -> FLATTEN -> FULLYCONNECTED\n",
" \n",
" Arguments:\n",
" X -- input dataset placeholder, of shape (input size, number of examples)\n",
" parameters -- python dictionary containing your parameters \"W1\", \"W2\"\n",
" the shapes are given in initialize_parameters\n",
"\n",
" Returns:\n",
" Z3 -- the output of the last LINEAR unit\n",
" \"\"\"\n",
" \n",
" # Retrieve the parameters from the dictionary \"parameters\" \n",
" W1 = parameters['W1']\n",
" W2 = parameters['W2']\n",
" \n",
" ### START CODE HERE ###\n",
" # CONV2D: stride of 1, padding 'SAME'\n",
" Z1 = tf.nn.conv2d(X, W1, strides=[1, 1, 1, 1], padding='SAME')\n",
" # RELU\n",
" A1 = tf.nn.relu(Z1)\n",
" # MAXPOOL: window 8x8, stride 8, padding 'SAME'\n",
" P1 = tf.nn.max_pool(A1, ksize = [1, 8, 8, 1], strides = [1, 8, 8, 1], padding='SAME')\n",
" # CONV2D: filters W2, stride 1, padding 'SAME'\n",
" Z2 = tf.nn.conv2d(P1, W2, strides=[1, 1, 1, 1], padding='SAME')\n",
" # RELU\n",
" A2 = tf.nn.relu(Z2)\n",
" # MAXPOOL: window 4x4, stride 4, padding 'SAME'\n",
" P2 = tf.nn.max_pool(A2, ksize = [1, 4, 4, 1], strides = [1, 4, 4, 1], padding='SAME')\n",
" # FLATTEN\n",
" P = tf.contrib.layers.flatten(P2)\n",
" # FULLY-CONNECTED without non-linear activation function (not not call softmax).\n",
" # 6 neurons in output layer. Hint: one of the arguments should be \"activation_fn=None\" \n",
" Z3 = tf.contrib.layers.fully_connected(P, 6, activation_fn=None)\n",
" ### END CODE HERE ###\n",
"\n",
" return Z3"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Z3 = [[-0.44670227 -1.57208765 -1.53049231 -2.31013036 -1.29104376 0.46852064]\n",
" [-0.17601591 -1.57972014 -1.4737016 -2.61672091 -1.00810647 0.5747785 ]]\n"
]
}
],
"source": [
"tf.reset_default_graph()\n",
"\n",
"with tf.Session() as sess:\n",
" np.random.seed(1)\n",
" X, Y = create_placeholders(64, 64, 3, 6)\n",
" parameters = initialize_parameters()\n",
" Z3 = forward_propagation(X, parameters)\n",
" init = tf.global_variables_initializer()\n",
" sess.run(init)\n",
" a = sess.run(Z3, {X: np.random.randn(2,64,64,3), Y: np.random.randn(2,6)})\n",
" print(\"Z3 = \" + str(a))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**:\n",
"\n",
"<table> \n",
" <td> \n",
" Z3 =\n",
" </td>\n",
" <td>\n",
" [[-0.44670227 -1.57208765 -1.53049231 -2.31013036 -1.29104376 0.46852064] <br>\n",
" [-0.17601591 -1.57972014 -1.4737016 -2.61672091 -1.00810647 0.5747785 ]]\n",
" </td>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 1.3 - Compute cost\n",
"\n",
"Implement the compute cost function below. You might find these two functions helpful: \n",
"\n",
"- **tf.nn.softmax_cross_entropy_with_logits(logits = Z3, labels = Y):** computes the softmax entropy loss. This function both computes the softmax activation function as well as the resulting loss. You can check the full documentation [here.](https://www.tensorflow.org/api_docs/python/tf/nn/softmax_cross_entropy_with_logits)\n",
"- **tf.reduce_mean:** computes the mean of elements across dimensions of a tensor. Use this to sum the losses over all the examples to get the overall cost. You can check the full documentation [here.](https://www.tensorflow.org/api_docs/python/tf/reduce_mean)\n",
"\n",
"** Exercise**: Compute the cost below using the function above."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: compute_cost \n",
"\n",
"def compute_cost(Z3, Y):\n",
" \"\"\"\n",
" Computes the cost\n",
" \n",
" Arguments:\n",
" Z3 -- output of forward propagation (output of the last LINEAR unit), of shape (6, number of examples)\n",
" Y -- \"true\" labels vector placeholder, same shape as Z3\n",
" \n",
" Returns:\n",
" cost - Tensor of the cost function\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ### (1 line of code)\n",
" cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=Z3, labels=Y))\n",
" ### END CODE HERE ###\n",
" \n",
" return cost"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"cost = 2.91034\n"
]
}
],
"source": [
"tf.reset_default_graph()\n",
"\n",
"with tf.Session() as sess:\n",
" np.random.seed(1)\n",
" X, Y = create_placeholders(64, 64, 3, 6)\n",
" parameters = initialize_parameters()\n",
" Z3 = forward_propagation(X, parameters)\n",
" cost = compute_cost(Z3, Y)\n",
" init = tf.global_variables_initializer()\n",
" sess.run(init)\n",
" a = sess.run(cost, {X: np.random.randn(4,64,64,3), Y: np.random.randn(4,6)})\n",
" print(\"cost = \" + str(a))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**: \n",
"\n",
"<table>\n",
" <td> \n",
" cost =\n",
" </td> \n",
" \n",
" <td> \n",
" 2.91034\n",
" </td> \n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1.4 Model \n",
"\n",
"Finally you will merge the helper functions you implemented above to build a model. You will train it on the SIGNS dataset. \n",
"\n",
"You have implemented `random_mini_batches()` in the Optimization programming assignment of course 2. Remember that this function returns a list of mini-batches. \n",
"\n",
"**Exercise**: Complete the function below. \n",
"\n",
"The model below should:\n",
"\n",
"- create placeholders\n",
"- initialize parameters\n",
"- forward propagate\n",
"- compute the cost\n",
"- create an optimizer\n",
"\n",
"Finally you will create a session and run a for loop for num_epochs, get the mini-batches, and then for each mini-batch you will optimize the function. [Hint for initializing the variables](https://www.tensorflow.org/api_docs/python/tf/global_variables_initializer)"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: model\n",
"\n",
"def model(X_train, Y_train, X_test, Y_test, learning_rate=0.009,\n",
" num_epochs=100, minibatch_size=64, print_cost=True):\n",
" \"\"\"\n",
" Implements a three-layer ConvNet in Tensorflow:\n",
" CONV2D -> RELU -> MAXPOOL -> CONV2D -> RELU -> MAXPOOL -> FLATTEN -> FULLYCONNECTED\n",
" \n",
" Arguments:\n",
" X_train -- training set, of shape (None, 64, 64, 3)\n",
" Y_train -- test set, of shape (None, n_y = 6)\n",
" X_test -- training set, of shape (None, 64, 64, 3)\n",
" Y_test -- test set, of shape (None, n_y = 6)\n",
" learning_rate -- learning rate of the optimization\n",
" num_epochs -- number of epochs of the optimization loop\n",
" minibatch_size -- size of a minibatch\n",
" print_cost -- True to print the cost every 100 epochs\n",
" \n",
" Returns:\n",
" train_accuracy -- real number, accuracy on the train set (X_train)\n",
" test_accuracy -- real number, testing accuracy on the test set (X_test)\n",
" parameters -- parameters learnt by the model. They can then be used to predict.\n",
" \"\"\"\n",
" \n",
" ops.reset_default_graph() # to be able to rerun the model without overwriting tf variables\n",
" tf.set_random_seed(1) # to keep results consistent (tensorflow seed)\n",
" seed = 3 # to keep results consistent (numpy seed)\n",
" (m, n_H0, n_W0, n_C0) = X_train.shape \n",
" n_y = Y_train.shape[1] \n",
" costs = [] # To keep track of the cost\n",
" \n",
" # Create Placeholders of the correct shape\n",
" ### START CODE HERE ### (1 line)\n",
" X, Y = create_placeholders(n_H0, n_W0, n_C0, n_y)\n",
" ### END CODE HERE ###\n",
"\n",
" # Initialize parameters\n",
" ### START CODE HERE ### (1 line)\n",
" parameters = initialize_parameters()\n",
" ### END CODE HERE ###\n",
" \n",
" # Forward propagation: Build the forward propagation in the tensorflow graph\n",
" ### START CODE HERE ### (1 line)\n",
" Z3 = forward_propagation(X, parameters)\n",
" ### END CODE HERE ###\n",
" \n",
" # Cost function: Add cost function to tensorflow graph\n",
" ### START CODE HERE ### (1 line)\n",
" cost = compute_cost(Z3, Y)\n",
" ### END CODE HERE ###\n",
" \n",
" # Backpropagation: Define the tensorflow optimizer. Use an AdamOptimizer that minimizes the cost.\n",
" ### START CODE HERE ### (1 line)\n",
" optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate).minimize(cost)\n",
" ### END CODE HERE ###\n",
" \n",
" # Initialize all the variables globally\n",
" init = tf.global_variables_initializer()\n",
" \n",
" # Start the session to compute the tensorflow graph\n",
" with tf.Session() as sess:\n",
" \n",
" # Run the initialization\n",
" sess.run(init)\n",
" \n",
" # Do the training loop\n",
" for epoch in range(num_epochs):\n",
"\n",
" minibatch_cost = 0.\n",
" num_minibatches = int(m / minibatch_size) # number of minibatches of size minibatch_size in the train set\n",
" seed = seed + 1\n",
" minibatches = random_mini_batches(X_train, Y_train, minibatch_size, seed)\n",
"\n",
" for minibatch in minibatches:\n",
"\n",
" # Select a minibatch\n",
" (minibatch_X, minibatch_Y) = minibatch\n",
" # IMPORTANT: The line that runs the graph on a minibatch.\n",
" # Run the session to execute the optimizer and the cost, the feedict should contain a minibatch for (X,Y).\n",
" ### START CODE HERE ### (1 line)\n",
" _ , temp_cost = sess.run([optimizer, cost], feed_dict={X:minibatch_X, Y:minibatch_Y})\n",
" ### END CODE HERE ###\n",
" \n",
" minibatch_cost += temp_cost / num_minibatches\n",
" \n",
"\n",
" # Print the cost every epoch\n",
" if print_cost == True and epoch % 5 == 0:\n",
" print (\"Cost after epoch %i: %f\" % (epoch, minibatch_cost))\n",
" if print_cost == True and epoch % 1 == 0:\n",
" costs.append(minibatch_cost)\n",
" \n",
" \n",
" # plot the cost\n",
" plt.plot(np.squeeze(costs))\n",
" plt.ylabel('cost')\n",
" plt.xlabel('iterations (per tens)')\n",
" plt.title(\"Learning rate =\" + str(learning_rate))\n",
" plt.show()\n",
"\n",
" # Calculate the correct predictions\n",
" predict_op = tf.argmax(Z3, 1)\n",
" correct_prediction = tf.equal(predict_op, tf.argmax(Y, 1))\n",
" \n",
" # Calculate accuracy on the test set\n",
" accuracy = tf.reduce_mean(tf.cast(correct_prediction, \"float\"))\n",
" print(accuracy)\n",
" train_accuracy = accuracy.eval({X: X_train, Y: Y_train})\n",
" test_accuracy = accuracy.eval({X: X_test, Y: Y_test})\n",
" print(\"Train Accuracy:\", train_accuracy)\n",
" print(\"Test Accuracy:\", test_accuracy)\n",
" \n",
" return train_accuracy, test_accuracy, parameters"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Run the following cell to train your model for 100 epochs. Check if your cost after epoch 0 and 5 matches our output. If not, stop the cell and go back to your code!"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Cost after epoch 0: 1.917929\n",
"Cost after epoch 5: 1.506757\n",
"Cost after epoch 10: 0.955359\n",
"Cost after epoch 15: 0.845802\n",
"Cost after epoch 20: 0.701174\n",
"Cost after epoch 25: 0.571977\n",
"Cost after epoch 30: 0.518435\n",
"Cost after epoch 35: 0.495806\n",
"Cost after epoch 40: 0.429827\n",
"Cost after epoch 45: 0.407291\n",
"Cost after epoch 50: 0.366394\n",
"Cost after epoch 55: 0.376922\n",
"Cost after epoch 60: 0.299491\n",
"Cost after epoch 65: 0.338870\n",
"Cost after epoch 70: 0.316400\n",
"Cost after epoch 75: 0.310413\n",
"Cost after epoch 80: 0.249549\n",
"Cost after epoch 85: 0.243457\n",
"Cost after epoch 90: 0.200031\n",
"Cost after epoch 95: 0.175452\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAEWCAYAAAB1xKBvAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nPypHYdmk//r2AAAAAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7fa860fcdc88>"
]
},
"metadata": {},
"output_type": "display_data"
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Tensor(\"Mean_1:0\", shape=(), dtype=float32)\n",
"Train Accuracy: 0.940741\n",
"Test Accuracy: 0.783333\n"
]
}
],
"source": [
"_, _, parameters = model(X_train, Y_train, X_test, Y_test)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected output**: although it may not match perfectly, your expected output should be close to ours and your cost value should decrease.\n",
"\n",
"<table> \n",
"<tr>\n",
" <td> \n",
" **Cost after epoch 0 =**\n",
" </td>\n",
"\n",
" <td> \n",
" 1.917929\n",
" </td> \n",
"</tr>\n",
"<tr>\n",
" <td> \n",
" **Cost after epoch 5 =**\n",
" </td>\n",
"\n",
" <td> \n",
" 1.506757\n",
" </td> \n",
"</tr>\n",
"<tr>\n",
" <td> \n",
" **Train Accuracy =**\n",
" </td>\n",
"\n",
" <td> \n",
" 0.940741\n",
" </td> \n",
"</tr> \n",
"\n",
"<tr>\n",
" <td> \n",
" **Test Accuracy =**\n",
" </td>\n",
"\n",
" <td> \n",
" 0.783333\n",
" </td> \n",
"</tr> \n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Congratulations! You have finised the assignment and built a model that recognizes SIGN language with almost 80% accuracy on the test set. If you wish, feel free to play around with this dataset further. You can actually improve its accuracy by spending more time tuning the hyperparameters, or using regularization (as this model clearly has a high variance). \n",
"\n",
"Once again, here's a thumbs up for your work! "
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.image.AxesImage at 0x7fa860df3588>"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7fa860edbfd0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fname = \"images/thumbs_up.jpg\"\n",
"image = np.array(ndimage.imread(fname, flatten=False))\n",
"my_image = scipy.misc.imresize(image, size=(64,64))\n",
"plt.imshow(my_image)"
]
}
],
"metadata": {
"coursera": {
"course_slug": "convolutional-neural-networks",
"graded_item_id": "bwbJV",
"launcher_item_id": "0TkXB"
},
"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": "3.6.0"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
================================================
FILE: Convolutional Neural Networks/Convolution model - Step by Step - v1.ipynb
================================================
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Convolutional Neural Networks: Step by Step\n",
"\n",
"Welcome to Course 4's first assignment! In this assignment, you will implement convolutional (CONV) and pooling (POOL) layers in numpy, including both forward propagation and (optionally) backward propagation. \n",
"\n",
"**Notation**:\n",
"- Superscript $[l]$ denotes an object of the $l^{th}$ layer. \n",
" - Example: $a^{[4]}$ is the $4^{th}$ layer activation. $W^{[5]}$ and $b^{[5]}$ are the $5^{th}$ layer parameters.\n",
"\n",
"\n",
"- Superscript $(i)$ denotes an object from the $i^{th}$ example. \n",
" - Example: $x^{(i)}$ is the $i^{th}$ training example input.\n",
" \n",
" \n",
"- Lowerscript $i$ denotes the $i^{th}$ entry of a vector.\n",
" - Example: $a^{[l]}_i$ denotes the $i^{th}$ entry of the activations in layer $l$, assuming this is a fully connected (FC) layer.\n",
" \n",
" \n",
"- $n_H$, $n_W$ and $n_C$ denote respectively the height, width and number of channels of a given layer. If you want to reference a specific layer $l$, you can also write $n_H^{[l]}$, $n_W^{[l]}$, $n_C^{[l]}$. \n",
"- $n_{H_{prev}}$, $n_{W_{prev}}$ and $n_{C_{prev}}$ denote respectively the height, width and number of channels of the previous layer. If referencing a specific layer $l$, this could also be denoted $n_H^{[l-1]}$, $n_W^{[l-1]}$, $n_C^{[l-1]}$. \n",
"\n",
"We assume that you are already familiar with `numpy` and/or have completed the previous courses of the specialization. Let's get started!"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1 - Packages\n",
"\n",
"Let's first import all the packages that you will need during this assignment. \n",
"- [numpy](www.numpy.org) is the fundamental package for scientific computing with Python.\n",
"- [matplotlib](http://matplotlib.org) is a library to plot graphs in Python.\n",
"- np.random.seed(1) is used to keep all the random function calls consistent. It will help us grade your work."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"import numpy as np\n",
"import h5py\n",
"import matplotlib.pyplot as plt\n",
"\n",
"%matplotlib inline\n",
"plt.rcParams['figure.figsize'] = (5.0, 4.0) # set default size of plots\n",
"plt.rcParams['image.interpolation'] = 'nearest'\n",
"plt.rcParams['image.cmap'] = 'gray'\n",
"\n",
"%load_ext autoreload\n",
"%autoreload 2\n",
"\n",
"np.random.seed(1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2 - Outline of the Assignment\n",
"\n",
"You will be implementing the building blocks of a convolutional neural network! Each function you will implement will have detailed instructions that will walk you through the steps needed:\n",
"\n",
"- Convolution functions, including:\n",
" - Zero Padding\n",
" - Convolve window \n",
" - Convolution forward\n",
" - Convolution backward (optional)\n",
"- Pooling functions, including:\n",
" - Pooling forward\n",
" - Create mask \n",
" - Distribute value\n",
" - Pooling backward (optional)\n",
" \n",
"This notebook will ask you to implement these functions from scratch in `numpy`. In the next notebook, you will use the TensorFlow equivalents of these functions to build the following model:\n",
"\n",
"<img src=\"images/model.png\" style=\"width:800px;height:300px;\">\n",
"\n",
"**Note** that for every forward function, there is its corresponding backward equivalent. Hence, at every step of your forward module you will store some parameters in a cache. These parameters are used to compute gradients during backpropagation. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3 - Convolutional Neural Networks\n",
"\n",
"Although programming frameworks make convolutions easy to use, they remain one of the hardest concepts to understand in Deep Learning. A convolution layer transforms an input volume into an output volume of different size, as shown below. \n",
"\n",
"<img src=\"images/conv_nn.png\" style=\"width:350px;height:200px;\">\n",
"\n",
"In this part, you will build every step of the convolution layer. You will first implement two helper functions: one for zero padding and the other for computing the convolution function itself. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3.1 - Zero-Padding\n",
"\n",
"Zero-padding adds zeros around the border of an image:\n",
"\n",
"<img src=\"images/PAD.png\" style=\"width:600px;height:400px;\">\n",
"<caption><center> <u> <font color='purple'> **Figure 1** </u><font color='purple'> : **Zero-Padding**<br> Image (3 channels, RGB) with a padding of 2. </center></caption>\n",
"\n",
"The main benefits of padding are the following:\n",
"\n",
"- It allows you to use a CONV layer without necessarily shrinking the height and width of the volumes. This is important for building deeper networks, since otherwise the height/width would shrink as you go to deeper layers. An important special case is the \"same\" convolution, in which the height/width is exactly preserved after one layer. \n",
"\n",
"- It helps us keep more of the information at the border of an image. Without padding, very few values at the next layer would be affected by pixels as the edges of an image.\n",
"\n",
"**Exercise**: Implement the following function, which pads all the images of a batch of examples X with zeros. [Use np.pad](https://docs.scipy.org/doc/numpy/reference/generated/numpy.pad.html). Note if you want to pad the array \"a\" of shape $(5,5,5,5,5)$ with `pad = 1` for the 2nd dimension, `pad = 3` for the 4th dimension and `pad = 0` for the rest, you would do:\n",
"```python\n",
"a = np.pad(a, ((0,0), (1,1), (0,0), (3,3), (0,0)), 'constant', constant_values = (..,..))\n",
"```"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: zero_pad\n",
"\n",
"def zero_pad(X, pad):\n",
" \"\"\"\n",
" Pad with zeros all images of the dataset X. The padding is applied to the height and width of an image, \n",
" as illustrated in Figure 1.\n",
" \n",
" Argument:\n",
" X -- python numpy array of shape (m, n_H, n_W, n_C) representing a batch of m images\n",
" pad -- integer, amount of padding around each image on vertical and horizontal dimensions\n",
" \n",
" Returns:\n",
" X_pad -- padded image of shape (m, n_H + 2*pad, n_W + 2*pad, n_C)\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ### (≈ 1 line)\n",
" X_pad = np.pad(X, ((0, 0), (pad, pad), (pad, pad), (0, 0)), 'constant', constant_values=0)\n",
" ### END CODE HERE ###\n",
" \n",
" return X_pad"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"x.shape = (4, 3, 3, 2)\n",
"x_pad.shape = (4, 7, 7, 2)\n",
"x[1, 1] = [[ 0.90085595 -0.68372786]\n",
" [-0.12289023 -0.93576943]\n",
" [-0.26788808 0.53035547]]\n",
"x_pad[1, 1] = [[ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]]\n"
]
},
{
"data": {
"text/plain": [
"<matplotlib.image.AxesImage at 0x7facfe2b51d0>"
]
},
"execution_count": 3,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAUgAAACuCAYAAABUfpQYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n0swshQPSzCyFA9LMLIUD0swsxf8DtS5DRn4HHEIAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7facfc68aa58>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"np.random.seed(1)\n",
"x = np.random.randn(4, 3, 3, 2)\n",
"x_pad = zero_pad(x, 2)\n",
"print (\"x.shape =\", x.shape)\n",
"print (\"x_pad.shape =\", x_pad.shape)\n",
"print (\"x[1, 1] =\", x[1, 1])\n",
"print (\"x_pad[1, 1] =\", x_pad[1, 1])\n",
"\n",
"fig, axarr = plt.subplots(1, 2)\n",
"axarr[0].set_title('x')\n",
"axarr[0].imshow(x[0,:,:,0])\n",
"axarr[1].set_title('x_pad')\n",
"axarr[1].imshow(x_pad[0,:,:,0])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**:\n",
"\n",
"<table>\n",
" <tr>\n",
" <td>\n",
" **x.shape**:\n",
" </td>\n",
" <td>\n",
" (4, 3, 3, 2)\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **x_pad.shape**:\n",
" </td>\n",
" <td>\n",
" (4, 7, 7, 2)\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **x[1,1]**:\n",
" </td>\n",
" <td>\n",
" [[ 0.90085595 -0.68372786]\n",
" [-0.12289023 -0.93576943]\n",
" [-0.26788808 0.53035547]]\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **x_pad[1,1]**:\n",
" </td>\n",
" <td>\n",
" [[ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]\n",
" [ 0. 0.]]\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 3.2 - Single step of convolution \n",
"\n",
"In this part, implement a single step of convolution, in which you apply the filter to a single position of the input. This will be used to build a convolutional unit, which: \n",
"\n",
"- Takes an input volume \n",
"- Applies a filter at every position of the input\n",
"- Outputs another volume (usually of different size)\n",
"\n",
"<img src=\"images/Convolution_schematic.gif\" style=\"width:500px;height:300px;\">\n",
"<caption><center> <u> <font color='purple'> **Figure 2** </u><font color='purple'> : **Convolution operation**<br> with a filter of 2x2 and a stride of 1 (stride = amount you move the window each time you slide) </center></caption>\n",
"\n",
"In a computer vision application, each value in the matrix on the left corresponds to a single pixel value, and we convolve a 3x3 filter with the image by multiplying its values element-wise with the original matrix, then summing them up. In this first step of the exercise, you will implement a single step of convolution, corresponding to applying a filter to just one of the positions to get a single real-valued output. \n",
"\n",
"Later in this notebook, you'll apply this function to multiple positions of the input to implement the full convolutional operation. \n",
"\n",
"**Exercise**: Implement conv_single_step(). [Hint](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.sum.html).\n"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: conv_single_step\n",
"\n",
"def conv_single_step(a_slice_prev, W, b):\n",
" \"\"\"\n",
" Apply one filter defined by parameters W on a single slice (a_slice_prev) of the output activation \n",
" of the previous layer.\n",
" \n",
" Arguments:\n",
" a_slice_prev -- slice of input data of shape (f, f, n_C_prev)\n",
" W -- Weight parameters contained in a window - matrix of shape (f, f, n_C_prev)\n",
" b -- Bias parameters contained in a window - matrix of shape (1, 1, 1)\n",
" \n",
" Returns:\n",
" Z -- a scalar value, result of convolving the sliding window (W, b) on a slice x of the input data\n",
" \"\"\"\n",
"\n",
" ### START CODE HERE ### (≈ 2 lines of code)\n",
" # Element-wise product between a_slice and W. Add bias.\n",
" s = np.multiply(a_slice_prev, W) + b\n",
" # Sum over all entries of the volume s\n",
" Z = np.sum(s)\n",
" ### END CODE HERE ###\n",
"\n",
" return Z"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Z = -23.1602122025\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"a_slice_prev = np.random.randn(4, 4, 3)\n",
"W = np.random.randn(4, 4, 3)\n",
"b = np.random.randn(1, 1, 1)\n",
"\n",
"Z = conv_single_step(a_slice_prev, W, b)\n",
"print(\"Z =\", Z)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**:\n",
"<table>\n",
" <tr>\n",
" <td>\n",
" **Z**\n",
" </td>\n",
" <td>\n",
" -23.1602122025\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"### 3.3 - Convolutional Neural Networks - Forward pass\n",
"\n",
"In the forward pass, you will take many filters and convolve them on the input. Each 'convolution' gives you a 2D matrix output. You will then stack these outputs to get a 3D volume: \n",
"\n",
"<center>\n",
"<video width=\"620\" height=\"440\" src=\"images/conv_kiank.mp4\" type=\"video/mp4\" controls>\n",
"</video>\n",
"</center>\n",
"\n",
"**Exercise**: Implement the function below to convolve the filters W on an input activation A_prev. This function takes as input A_prev, the activations output by the previous layer (for a batch of m inputs), F filters/weights denoted by W, and a bias vector denoted by b, where each filter has its own (single) bias. Finally you also have access to the hyperparameters dictionary which contains the stride and the padding. \n",
"\n",
"**Hint**: \n",
"1. To select a 2x2 slice at the upper left corner of a matrix \"a_prev\" (shape (5,5,3)), you would do:\n",
"```python\n",
"a_slice_prev = a_prev[0:2,0:2,:]\n",
"```\n",
"This will be useful when you will define `a_slice_prev` below, using the `start/end` indexes you will define.\n",
"2. To define a_slice you will need to first define its corners `vert_start`, `vert_end`, `horiz_start` and `horiz_end`. This figure may be helpful for you to find how each of the corner can be defined using h, w, f and s in the code below.\n",
"\n",
"<img src=\"images/vert_horiz_kiank.png\" style=\"width:400px;height:300px;\">\n",
"<caption><center> <u> <font color='purple'> **Figure 3** </u><font color='purple'> : **Definition of a slice using vertical and horizontal start/end (with a 2x2 filter)** <br> This figure shows only a single channel. </center></caption>\n",
"\n",
"\n",
"**Reminder**:\n",
"The formulas relating the output shape of the convolution to the input shape is:\n",
"$$ n_H = \\lfloor \\frac{n_{H_{prev}} - f + 2 \\times pad}{stride} \\rfloor +1 $$\n",
"$$ n_W = \\lfloor \\frac{n_{W_{prev}} - f + 2 \\times pad}{stride} \\rfloor +1 $$\n",
"$$ n_C = \\text{number of filters used in the convolution}$$\n",
"\n",
"For this exercise, we won't worry about vectorization, and will just implement everything with for-loops."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: conv_forward\n",
"\n",
"def conv_forward(A_prev, W, b, hparameters):\n",
" \"\"\"\n",
" Implements the forward propagation for a convolution function\n",
" \n",
" Arguments:\n",
" A_prev -- output activations of the previous layer, numpy array of shape (m, n_H_prev, n_W_prev, n_C_prev)\n",
" W -- Weights, numpy array of shape (f, f, n_C_prev, n_C)\n",
" b -- Biases, numpy array of shape (1, 1, 1, n_C)\n",
" hparameters -- python dictionary containing \"stride\" and \"pad\"\n",
" \n",
" Returns:\n",
" Z -- conv output, numpy array of shape (m, n_H, n_W, n_C)\n",
" cache -- cache of values needed for the conv_backward() function\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ###\n",
" # Retrieve dimensions from A_prev's shape (≈1 line) \n",
" (m, n_H_prev, n_W_prev, n_C_prev) = A_prev.shape\n",
" \n",
" # Retrieve dimensions from W's shape (≈1 line)\n",
" (f, f, n_C_prev, n_C) = W.shape\n",
"\n",
" # Retrieve information from \"hparameters\" (≈2 lines)\n",
" stride = hparameters['stride']\n",
" pad = hparameters['pad']\n",
" \n",
" # Compute the dimensions of the CONV output volume using the formula given above. Hint: use int() to floor. (≈2 lines)\n",
" n_H = int((n_H_prev - f + 2 * pad) / stride) + 1\n",
" n_W = int((n_W_prev - f + 2 * pad) / stride) + 1\n",
" \n",
" # Initialize the output volume Z with zeros. (≈1 line)\n",
" Z = np.zeros((m, n_H, n_W, n_C))\n",
" \n",
" # Create A_prev_pad by padding A_prev\n",
" A_prev_pad = zero_pad(A_prev, pad)\n",
" \n",
" for i in range(m): # loop over the batch of training examples\n",
" a_prev_pad = A_prev_pad[i] # Select ith training example's padded activation\n",
" for h in range(n_H): # loop over vertical axis of the output volume\n",
" for w in range(n_W): # loop over horizontal axis of the output volume\n",
" for c in range(n_C): # loop over channels (= #filters) of the output volume\n",
" # Find the corners of the current \"slice\" (≈4 lines)\n",
" vert_start = h * stride\n",
" vert_end = vert_start + f\n",
" horiz_start = w * stride\n",
" horiz_end = horiz_start + f\n",
" # Use the corners to define the (3D) slice of a_prev_pad (See Hint above the cell). (≈1 line)\n",
" a_slice_prev = a_prev_pad[vert_start:vert_end, horiz_start:horiz_end, :]\n",
" # Convolve the (3D) slice with the correct filter W and bias b, to get back one output neuron. (≈1 line)\n",
" Z[i, h, w, c] = conv_single_step(a_slice_prev, W[...,c], b[...,c])\n",
" \n",
" ### END CODE HERE ###\n",
"\n",
" # Making sure your output shape is correct\n",
" assert(Z.shape == (m, n_H, n_W, n_C))\n",
" \n",
" # Save information in \"cache\" for the backprop\n",
" cache = (A_prev, W, b, hparameters)\n",
" \n",
" return Z, cache"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Z's mean = 0.155859324889\n",
"cache_conv[0][1][2][3] = [-0.20075807 0.18656139 0.41005165]\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"A_prev = np.random.randn(10, 4, 4, 3)\n",
"W = np.random.randn(2, 2, 3, 8)\n",
"b = np.random.randn(1, 1, 1, 8)\n",
"hparameters = {\"pad\" : 2,\n",
" \"stride\": 1}\n",
"\n",
"Z, cache_conv = conv_forward(A_prev, W, b, hparameters)\n",
"print(\"Z's mean =\", np.mean(Z))\n",
"print(\"cache_conv[0][1][2][3] =\", cache_conv[0][1][2][3])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**:\n",
"\n",
"<table>\n",
" <tr>\n",
" <td>\n",
" **Z's mean**\n",
" </td>\n",
" <td>\n",
" 0.155859324889\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **cache_conv[0][1][2][3]**\n",
" </td>\n",
" <td>\n",
" [-0.20075807 0.18656139 0.41005165]\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Finally, CONV layer should also contain an activation, in which case we would add the following line of code:\n",
"\n",
"```python\n",
"# Convolve the window to get back one output neuron\n",
"Z[i, h, w, c] = ...\n",
"# Apply activation\n",
"A[i, h, w, c] = activation(Z[i, h, w, c])\n",
"```\n",
"\n",
"You don't need to do it here. \n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4 - Pooling layer \n",
"\n",
"The pooling (POOL) layer reduces the height and width of the input. It helps reduce computation, as well as helps make feature detectors more invariant to its position in the input. The two types of pooling layers are: \n",
"\n",
"- Max-pooling layer: slides an ($f, f$) window over the input and stores the max value of the window in the output.\n",
"\n",
"- Average-pooling layer: slides an ($f, f$) window over the input and stores the average value of the window in the output.\n",
"\n",
"<table>\n",
"<td>\n",
"<img src=\"images/max_pool1.png\" style=\"width:500px;height:300px;\">\n",
"<td>\n",
"\n",
"<td>\n",
"<img src=\"images/a_pool.png\" style=\"width:500px;height:300px;\">\n",
"<td>\n",
"</table>\n",
"\n",
"These pooling layers have no parameters for backpropagation to train. However, they have hyperparameters such as the window size $f$. This specifies the height and width of the fxf window you would compute a max or average over. \n",
"\n",
"### 4.1 - Forward Pooling\n",
"Now, you are going to implement MAX-POOL and AVG-POOL, in the same function. \n",
"\n",
"**Exercise**: Implement the forward pass of the pooling layer. Follow the hints in the comments below.\n",
"\n",
"**Reminder**:\n",
"As there's no padding, the formulas binding the output shape of the pooling to the input shape is:\n",
"$$ n_H = \\lfloor \\frac{n_{H_{prev}} - f}{stride} \\rfloor +1 $$\n",
"$$ n_W = \\lfloor \\frac{n_{W_{prev}} - f}{stride} \\rfloor +1 $$\n",
"$$ n_C = n_{C_{prev}}$$"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: pool_forward\n",
"\n",
"def pool_forward(A_prev, hparameters, mode = \"max\"):\n",
" \"\"\"\n",
" Implements the forward pass of the pooling layer\n",
" \n",
" Arguments:\n",
" A_prev -- Input data, numpy array of shape (m, n_H_prev, n_W_prev, n_C_prev)\n",
" hparameters -- python dictionary containing \"f\" and \"stride\"\n",
" mode -- the pooling mode you would like to use, defined as a string (\"max\" or \"average\")\n",
" \n",
" Returns:\n",
" A -- output of the pool layer, a numpy array of shape (m, n_H, n_W, n_C)\n",
" cache -- cache used in the backward pass of the pooling layer, contains the input and hparameters \n",
" \"\"\"\n",
" \n",
" # Retrieve dimensions from the input shape\n",
" (m, n_H_prev, n_W_prev, n_C_prev) = A_prev.shape\n",
" \n",
" # Retrieve hyperparameters from \"hparameters\"\n",
" f = hparameters[\"f\"]\n",
" stride = hparameters[\"stride\"]\n",
" \n",
" # Define the dimensions of the output\n",
" n_H = int(1 + (n_H_prev - f) / stride)\n",
" n_W = int(1 + (n_W_prev - f) / stride)\n",
" n_C = n_C_prev\n",
" \n",
" # Initialize output matrix A\n",
" A = np.zeros((m, n_H, n_W, n_C)) \n",
" \n",
" ### START CODE HERE ###\n",
" for i in range(m): # loop over the training examples\n",
" for h in range(n_H): # loop on the vertical axis of the output volume\n",
" for w in range(n_W): # loop on the horizontal axis of the output volume\n",
" for c in range (n_C): # loop over the channels of the output volume\n",
" \n",
" # Find the corners of the current \"slice\" (≈4 lines)\n",
" vert_start = h * stride\n",
" vert_end = vert_start + f\n",
" horiz_start = w * stride\n",
" horiz_end = horiz_start + f\n",
" \n",
" # Use the corners to define the current slice on the ith training example of A_prev, channel c. (≈1 line)\n",
" a_prev_slice = A_prev[i, vert_start:vert_end, horiz_start:horiz_end, c]\n",
" \n",
" # Compute the pooling operation on the slice. Use an if statment to differentiate the modes. Use np.max/np.mean.\n",
" if mode == \"max\":\n",
" A[i, h, w, c] = np.max(a_prev_slice)\n",
" elif mode == \"average\":\n",
" A[i, h, w, c] = np.mean(a_prev_slice)\n",
" \n",
" ### END CODE HERE ###\n",
" \n",
" # Store the input and hparameters in \"cache\" for pool_backward()\n",
" cache = (A_prev, hparameters)\n",
" \n",
" # Making sure your output shape is correct\n",
" assert(A.shape == (m, n_H, n_W, n_C))\n",
" \n",
" return A, cache\n"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"mode = max\n",
"A = [[[[ 1.74481176 1.6924546 2.10025514]]]\n",
"\n",
"\n",
" [[[ 1.19891788 1.51981682 2.18557541]]]]\n",
"\n",
"mode = average\n",
"A = [[[[-0.09498456 0.11180064 -0.14263511]]]\n",
"\n",
"\n",
" [[[-0.09525108 0.28325018 0.33035185]]]]\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"A_prev = np.random.randn(2, 4, 4, 3)\n",
"hparameters = {\"stride\" : 1, \"f\": 4}\n",
"\n",
"A, cache = pool_forward(A_prev, hparameters)\n",
"print(\"mode = max\")\n",
"print(\"A =\", A)\n",
"print()\n",
"A, cache = pool_forward(A_prev, hparameters, mode = \"average\")\n",
"print(\"mode = average\")\n",
"print(\"A =\", A)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output:**\n",
"<table>\n",
"\n",
" <tr>\n",
" <td>\n",
" A =\n",
" </td>\n",
" <td>\n",
" [[[[ 1.74481176 1.6924546 2.10025514]]] <br/>\n",
"\n",
"\n",
" [[[ 1.19891788 1.51981682 2.18557541]]]]\n",
"\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" A =\n",
" </td>\n",
" <td>\n",
" [[[[-0.09498456 0.11180064 -0.14263511]]] <br/>\n",
"\n",
"\n",
" [[[-0.09525108 0.28325018 0.33035185]]]]\n",
"\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Congratulations! You have now implemented the forward passes of all the layers of a convolutional network. \n",
"\n",
"The remainer of this notebook is optional, and will not be graded.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 5 - Backpropagation in convolutional neural networks (OPTIONAL / UNGRADED)\n",
"\n",
"In modern deep learning frameworks, you only have to implement the forward pass, and the framework takes care of the backward pass, so most deep learning engineers don't need to bother with the details of the backward pass. The backward pass for convolutional networks is complicated. If you wish however, you can work through this optional portion of the notebook to get a sense of what backprop in a convolutional network looks like. \n",
"\n",
"When in an earlier course you implemented a simple (fully connected) neural network, you used backpropagation to compute the derivatives with respect to the cost to update the parameters. Similarly, in convolutional neural networks you can to calculate the derivatives with respect to the cost in order to update the parameters. The backprop equations are not trivial and we did not derive them in lecture, but we briefly presented them below.\n",
"\n",
"### 5.1 - Convolutional layer backward pass \n",
"\n",
"Let's start by implementing the backward pass for a CONV layer. \n",
"\n",
"#### 5.1.1 - Computing dA:\n",
"This is the formula for computing $dA$ with respect to the cost for a certain filter $W_c$ and a given training example:\n",
"\n",
"$$ dA += \\sum _{h=0} ^{n_H} \\sum_{w=0} ^{n_W} W_c \\times dZ_{hw} \\tag{1}$$\n",
"\n",
"Where $W_c$ is a filter and $dZ_{hw}$ is a scalar corresponding to the gradient of the cost with respect to the output of the conv layer Z at the hth row and wth column (corresponding to the dot product taken at the ith stride left and jth stride down). Note that at each time, we multiply the the same filter $W_c$ by a different dZ when updating dA. We do so mainly because when computing the forward propagation, each filter is dotted and summed by a different a_slice. Therefore when computing the backprop for dA, we are just adding the gradients of all the a_slices. \n",
"\n",
"In code, inside the appropriate for-loops, this formula translates into:\n",
"```python\n",
"da_prev_pad[vert_start:vert_end, horiz_start:horiz_end, :] += W[:,:,:,c] * dZ[i, h, w, c]\n",
"```\n",
"\n",
"#### 5.1.2 - Computing dW:\n",
"This is the formula for computing $dW_c$ ($dW_c$ is the derivative of one filter) with respect to the loss:\n",
"\n",
"$$ dW_c += \\sum _{h=0} ^{n_H} \\sum_{w=0} ^ {n_W} a_{slice} \\times dZ_{hw} \\tag{2}$$\n",
"\n",
"Where $a_{slice}$ corresponds to the slice which was used to generate the acitivation $Z_{ij}$. Hence, this ends up giving us the gradient for $W$ with respect to that slice. Since it is the same $W$, we will just add up all such gradients to get $dW$. \n",
"\n",
"In code, inside the appropriate for-loops, this formula translates into:\n",
"```python\n",
"dW[:,:,:,c] += a_slice * dZ[i, h, w, c]\n",
"```\n",
"\n",
"#### 5.1.3 - Computing db:\n",
"\n",
"This is the formula for computing $db$ with respect to the cost for a certain filter $W_c$:\n",
"\n",
"$$ db = \\sum_h \\sum_w dZ_{hw} \\tag{3}$$\n",
"\n",
"As you have previously seen in basic neural networks, db is computed by summing $dZ$. In this case, you are just summing over all the gradients of the conv output (Z) with respect to the cost. \n",
"\n",
"In code, inside the appropriate for-loops, this formula translates into:\n",
"```python\n",
"db[:,:,:,c] += dZ[i, h, w, c]\n",
"```\n",
"\n",
"**Exercise**: Implement the `conv_backward` function below. You should sum over all the training examples, filters, heights, and widths. You should then compute the derivatives using formulas 1, 2 and 3 above. "
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def conv_backward(dZ, cache):\n",
" \"\"\"\n",
" Implement the backward propagation for a convolution function\n",
" \n",
" Arguments:\n",
" dZ -- gradient of the cost with respect to the output of the conv layer (Z), numpy array of shape (m, n_H, n_W, n_C)\n",
" cache -- cache of values needed for the conv_backward(), output of conv_forward()\n",
" \n",
" Returns:\n",
" dA_prev -- gradient of the cost with respect to the input of the conv layer (A_prev),\n",
" numpy array of shape (m, n_H_prev, n_W_prev, n_C_prev)\n",
" dW -- gradient of the cost with respect to the weights of the conv layer (W)\n",
" numpy array of shape (f, f, n_C_prev, n_C)\n",
" db -- gradient of the cost with respect to the biases of the conv layer (b)\n",
" numpy array of shape (1, 1, 1, n_C)\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ###\n",
" # Retrieve information from \"cache\"\n",
" (A_prev, W, b, hparameters) = cache\n",
" \n",
" # Retrieve dimensions from A_prev's shape\n",
" (m, n_H_prev, n_W_prev, n_C_prev) = A_prev.shape\n",
" \n",
" # Retrieve dimensions from W's shape\n",
" (f, f, n_C_prev, n_C) = W.shape\n",
" \n",
" # Retrieve information from \"hparameters\"\n",
" stride = hparameters[\"stride\"]\n",
" pad = hparameters[\"pad\"]\n",
" \n",
" # Retrieve dimensions from dZ's shape\n",
" (m, n_H, n_W, n_C) = dZ.shape\n",
" \n",
" # Initialize dA_prev, dW, db with the correct shapes\n",
" dA_prev = np.zeros((m, n_H_prev, n_W_prev, n_C_prev)) \n",
" dW = np.zeros((f, f, n_C_prev, n_C))\n",
" db = np.zeros((1, 1, 1, n_C))\n",
"\n",
" # Pad A_prev and dA_prev\n",
" A_prev_pad = zero_pad(A_prev, pad)\n",
" dA_prev_pad = zero_pad(dA_prev, pad)\n",
" \n",
" for i in range(m): # loop over the training examples\n",
" \n",
" # select ith training example from A_prev_pad and dA_prev_pad\n",
" a_prev_pad = A_prev_pad[i]\n",
" da_prev_pad = dA_prev_pad[i]\n",
" \n",
" for h in range(n_H): # loop over vertical axis of the output volume\n",
" for w in range(n_W): # loop over horizontal axis of the output volume\n",
" for c in range(n_C): # loop over the channels of the output volume\n",
" \n",
" # Find the corners of the current \"slice\"\n",
" vert_start = h * stride\n\n",
" vert_end = vert_start + f\n",
" horiz_start = w * stride\n\n",
" horiz_end = horiz_start + f\n",
" \n",
" # Use the corners to define the slice from a_prev_pad\n",
" a_slice = a_prev_pad[vert_start:vert_end, horiz_start:horiz_end, :]\n",
"\n",
" # Update gradients for the window and the filter's parameters using the code formulas given above\n",
" da_prev_pad[vert_start:vert_end, horiz_start:horiz_end, :] += W[:,:,:,c] * dZ[i, h, w, c]\n",
" dW[:,:,:,c] += a_slice * dZ[i, h, w, c]\n",
" db[:,:,:,c] += dZ[i, h, w, c]\n",
" \n",
" # Set the ith training example's dA_prev to the unpaded da_prev_pad (Hint: use X[pad:-pad, pad:-pad, :])\n",
" dA_prev[i, :, :, :] = da_prev_pad[pad:-pad, pad:-pad, :]\n",
" ### END CODE HERE ###\n",
" \n",
" # Making sure your output shape is correct\n",
" assert(dA_prev.shape == (m, n_H_prev, n_W_prev, n_C_prev))\n",
" \n",
" return dA_prev, dW, db"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"dA_mean = 9.60899067587\n",
"dW_mean = 10.5817412755\n",
"db_mean = 76.3710691956\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"dA, dW, db = conv_backward(Z, cache_conv)\n",
"print(\"dA_mean =\", np.mean(dA))\n",
"print(\"dW_mean =\", np.mean(dW))\n",
"print(\"db_mean =\", np.mean(db))\n",
"# print(dA.shape)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"** Expected Output: **\n",
"<table>\n",
" <tr>\n",
" <td>\n",
" **dA_mean**\n",
" </td>\n",
" <td>\n",
" 9.60899067587\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **dW_mean**\n",
" </td>\n",
" <td>\n",
" 10.5817412755\n",
" </td>\n",
" </tr>\n",
" <tr>\n",
" <td>\n",
" **db_mean**\n",
" </td>\n",
" <td>\n",
" 76.3710691956\n",
" </td>\n",
" </tr>\n",
"\n",
"</table>\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 5.2 Pooling layer - backward pass\n",
"\n",
"Next, let's implement the backward pass for the pooling layer, starting with the MAX-POOL layer. Even though a pooling layer has no parameters for backprop to update, you still need to backpropagation the gradient through the pooling layer in order to compute gradients for layers that came before the pooling layer. \n",
"\n",
"### 5.2.1 Max pooling - backward pass \n",
"\n",
"Before jumping into the backpropagation of the pooling layer, you are going to build a helper function called `create_mask_from_window()` which does the following: \n",
"\n",
"$$ X = \\begin{bmatrix}\n",
"1 && 3 \\\\\n",
"4 && 2\n",
"\\end{bmatrix} \\quad \\rightarrow \\quad M =\\begin{bmatrix}\n",
"0 && 0 \\\\\n",
"1 && 0\n",
"\\end{bmatrix}\\tag{4}$$\n",
"\n",
"As you can see, this function creates a \"mask\" matrix which keeps track of where the maximum of the matrix is. True (1) indicates the position of the maximum in X, the other entries are False (0). You'll see later that the backward pass for average pooling will be similar to this but using a different mask. \n",
"\n",
"**Exercise**: Implement `create_mask_from_window()`. This function will be helpful for pooling backward. \n",
"Hints:\n",
"- [np.max()]() may be helpful. It computes the maximum of an array.\n",
"- If you have a matrix X and a scalar x: `A = (X == x)` will return a matrix A of the same size as X such that:\n",
"```\n",
"A[i,j] = True if X[i,j] = x\n",
"A[i,j] = False if X[i,j] != x\n",
"```\n",
"- Here, you don't need to consider cases where there are several maxima in a matrix."
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def create_mask_from_window(x):\n",
" \"\"\"\n",
" Creates a mask from an input matrix x, to identify the max entry of x.\n",
" \n",
" Arguments:\n",
" x -- Array of shape (f, f)\n",
" \n",
" Returns:\n",
" mask -- Array of the same shape as window, contains a True at the position corresponding to the max entry of x.\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ### (≈1 line)\n",
" mask = x == np.max(x)\n",
" ### END CODE HERE ###\n",
" \n",
" return mask"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"x = [[ 1.62434536 -0.61175641 -0.52817175]\n",
" [-1.07296862 0.86540763 -2.3015387 ]]\n",
"mask = [[ True False False]\n",
" [False False False]]\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"x = np.random.randn(2,3)\n",
"mask = create_mask_from_window(x)\n",
"print('x = ', x)\n",
"print(\"mask = \", mask)"
]
},
{
"cell_type": "markdown",
"metadata": {
"collapsed": true
},
"source": [
"**Expected Output:** \n",
"\n",
"<table> \n",
"<tr> \n",
"<td>\n",
"\n",
"**x =**\n",
"</td>\n",
"\n",
"<td>\n",
"\n",
"[[ 1.62434536 -0.61175641 -0.52817175] <br>\n",
" [-1.07296862 0.86540763 -2.3015387 ]]\n",
"\n",
" </td>\n",
"</tr>\n",
"\n",
"<tr> \n",
"<td>\n",
"**mask =**\n",
"</td>\n",
"<td>\n",
"[[ True False False] <br>\n",
" [False False False]]\n",
"</td>\n",
"</tr>\n",
"\n",
"\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Why do we keep track of the position of the max? It's because this is the input value that ultimately influenced the output, and therefore the cost. Backprop is computing gradients with respect to the cost, so anything that influences the ultimate cost should have a non-zero gradient. So, backprop will \"propagate\" the gradient back to this particular input value that had influenced the cost. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 5.2.2 - Average pooling - backward pass \n",
"\n",
"In max pooling, for each input window, all the \"influence\" on the output came from a single input value--the max. In average pooling, every element of the input window has equal influence on the output. So to implement backprop, you will now implement a helper function that reflects this.\n",
"\n",
"For example if we did average pooling in the forward pass using a 2x2 filter, then the mask you'll use for the backward pass will look like: \n",
"$$ dZ = 1 \\quad \\rightarrow \\quad dZ =\\begin{bmatrix}\n",
"1/4 && 1/4 \\\\\n",
"1/4 && 1/4\n",
"\\end{bmatrix}\\tag{5}$$\n",
"\n",
"This implies that each position in the $dZ$ matrix contributes equally to output because in the forward pass, we took an average. \n",
"\n",
"**Exercise**: Implement the function below to equally distribute a value dz through a matrix of dimension shape. [Hint](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.ones.html)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def distribute_value(dz, shape):\n",
" \"\"\"\n",
" Distributes the input value in the matrix of dimension shape\n",
" \n",
" Arguments:\n",
" dz -- input scalar\n",
" shape -- the shape (n_H, n_W) of the output matrix for which we want to distribute the value of dz\n",
" \n",
" Returns:\n",
" a -- Array of size (n_H, n_W) for which we distributed the value of dz\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ###\n",
" # Retrieve dimensions from shape (≈1 line)\n",
" (n_H, n_W) = shape\n",
" \n",
" # Compute the value to distribute on the matrix (≈1 line)\n",
" average = dz / (n_H * n_W)\n",
" \n",
" # Create a matrix where every entry is the \"average\" value (≈1 line)\n",
" a = np.ones(shape) * average\n",
" ### END CODE HERE ###\n",
" \n",
" return a"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"distributed value = [[ 0.5 0.5]\n",
" [ 0.5 0.5]]\n"
]
}
],
"source": [
"a = distribute_value(2, (2,2))\n",
"print('distributed value =', a)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**: \n",
"\n",
"<table> \n",
"<tr> \n",
"<td>\n",
"distributed_value =\n",
"</td>\n",
"<td>\n",
"[[ 0.5 0.5]\n",
"<br\\> \n",
"[ 0.5 0.5]]\n",
"</td>\n",
"</tr>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### 5.2.3 Putting it together: Pooling backward \n",
"\n",
"You now have everything you need to compute backward propagation on a pooling layer.\n",
"\n",
"**Exercise**: Implement the `pool_backward` function in both modes (`\"max\"` and `\"average\"`). You will once again use 4 for-loops (iterating over training examples, height, width, and channels). You should use an `if/elif` statement to see if the mode is equal to `'max'` or `'average'`. If it is equal to 'average' you should use the `distribute_value()` function you implemented above to create a matrix of the same shape as `a_slice`. Otherwise, the mode is equal to '`max`', and you will create a mask with `create_mask_from_window()` and multiply it by the corresponding value of dZ."
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"\n",
"def pool_backward(dA, cache, mode = \"max\"):\n",
" \"\"\"\n",
" Implements the backward pass of the pooling layer\n",
" \n",
" Arguments:\n",
" dA -- gradient of cost with respect to the output of the pooling layer, same shape as A\n",
" cache -- cache output from the forward pass of the pooling layer, contains the layer's input and hparameters \n",
" mode -- the pooling mode you would like to use, defined as a string (\"max\" or \"average\")\n",
" \n",
" Returns:\n",
" dA_prev -- gradient of cost with respect to the input of the pooling layer, same shape as A_prev\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ###\n",
" \n",
" # Retrieve information from cache (≈1 line)\n",
" (A_prev, hparameters) = cache\n",
" \n",
" # Retrieve hyperparameters from \"hparameters\" (≈2 lines)\n",
" stride = hparameters[\"stride\"]\n",
" f = hparameters[\"f\"]\n",
" \n",
" # Retrieve dimensions from A_prev's shape and dA's shape (≈2 lines)\n",
" m, n_H_prev, n_W_prev, n_C_prev = A_prev.shape\n",
" m, n_H, n_W, n_C = dA.shape\n",
" \n",
" # Initialize dA_prev with zeros (≈1 line)\n",
" dA_prev = np.zeros(A_prev.shape)\n",
" \n",
" for i in range(m): # loop over the training examples\n",
" # select training example from A_prev (≈1 line)\n",
" a_prev = A_prev[i]\n",
" for h in range(n_H): # loop on the vertical axis\n",
" for w in range(n_W): # loop on the horizontal axis\n",
" for c in range(n_C): # loop over the channels (depth)\n",
" # Find the corners of the current \"slice\" (≈4 lines)\n",
" vert_start = h\n",
" vert_end = vert_start + f\n",
" horiz_start = w\n",
" horiz_end = horiz_start + f\n",
" \n",
" # Compute the backward propagation in both modes.\n",
" if mode == \"max\":\n",
" # Use the corners and \"c\" to define the current slice from a_prev (≈1 line)\n",
" a_prev_slice = a_prev[vert_start:vert_end, horiz_start:horiz_end, c]\n",
" # Create the mask from a_prev_slice (≈1 line)\n",
" mask = create_mask_from_window(a_prev_slice)\n",
" # Set dA_prev to be dA_prev + (the mask multiplied by the correct entry of dA) (≈1 line)\n",
" dA_prev[i, vert_start:vert_end, horiz_start:horiz_end, c] += np.multiply(mask, dA[i, h, w, c])\n",
" \n",
" elif mode == \"average\":\n",
" # Get the value a from dA (≈1 line)\n",
" da = dA[i, h, w, c]\n",
" # Define the shape of the filter as fxf (≈1 line)\n",
" shape = (f, f)\n",
" # Distribute it to get the correct slice of dA_prev. i.e. Add the distributed value of da. (≈1 line)\n",
" dA_prev[i, vert_start:vert_end, horiz_start:horiz_end, c] += distribute_value(da, shape)\n",
" \n",
" ### END CODE ###\n",
" \n",
" # Making sure your output shape is correct\n",
" assert(dA_prev.shape == A_prev.shape)\n",
" \n",
" return dA_prev"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"mode = max\n",
"mean of dA = 0.145713902729\n",
"dA_prev[1,1] = [[ 0. 0. ]\n",
" [ 5.05844394 -1.68282702]\n",
" [ 0. 0. ]]\n",
"\n",
"mode = average\n",
"mean of dA = 0.145713902729\n",
"dA_prev[1,1] = [[ 0.08485462 0.2787552 ]\n",
" [ 1.26461098 -0.25749373]\n",
" [ 1.17975636 -0.53624893]]\n"
]
}
],
"source": [
"np.random.seed(1)\n",
"A_prev = np.random.randn(5, 5, 3, 2)\n",
"hparameters = {\"stride\" : 1, \"f\": 2}\n",
"A, cache = pool_forward(A_prev, hparameters)\n",
"dA = np.random.randn(5, 4, 2, 2)\n",
"\n",
"dA_prev = pool_backward(dA, cache, mode = \"max\")\n",
"print(\"mode = max\")\n",
"print('mean of dA = ', np.mean(dA))\n",
"print('dA_prev[1,1] = ', dA_prev[1,1]) \n",
"print()\n",
"dA_prev = pool_backward(dA, cache, mode = \"average\")\n",
"print(\"mode = average\")\n",
"print('mean of dA = ', np.mean(dA))\n",
"print('dA_prev[1,1] = ', dA_prev[1,1]) "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Expected Output**: \n",
"\n",
"mode = max:\n",
"<table> \n",
"<tr> \n",
"<td>\n",
"\n",
"**mean of dA =**\n",
"</td>\n",
"\n",
"<td>\n",
"\n",
"0.145713902729\n",
"\n",
" </td>\n",
"</tr>\n",
"\n",
"<tr> \n",
"<td>\n",
"**dA_prev[1,1] =** \n",
"</td>\n",
"<td>\n",
"[[ 0. 0. ] <br>\n",
" [ 5.05844394 -1.68282702] <br>\n",
" [ 0. 0. ]]\n",
"</td>\n",
"</tr>\n",
"</table>\n",
"\n",
"mode = average\n",
"<table> \n",
"<tr> \n",
"<td>\n",
"\n",
"**mean of dA =**\n",
"</td>\n",
"\n",
"<td>\n",
"\n",
"0.145713902729\n",
"\n",
" </td>\n",
"</tr>\n",
"\n",
"<tr> \n",
"<td>\n",
"**dA_prev[1,1] =** \n",
"</td>\n",
"<td>\n",
"[[ 0.08485462 0.2787552 ] <br>\n",
" [ 1.26461098 -0.25749373] <br>\n",
" [ 1.17975636 -0.53624893]]\n",
"</td>\n",
"</tr>\n",
"</table>"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Congratulations !\n",
"\n",
"Congratulation on completing this assignment. You now understand how convolutional neural networks work. You have implemented all the building blocks of a neural network. In the next assignment you will implement a ConvNet using TensorFlow."
]
}
],
"metadata": {
"coursera": {
"course_slug": "convolutional-neural-networks",
"graded_item_id": "qO8ng",
"launcher_item_id": "7XDi8"
},
"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": "3.5.3"
}
},
"nbformat": 4,
"nbformat_minor": 1
}
================================================
FILE: Convolutional Neural Networks/Keras - Tutorial - Happy House v1.ipynb
================================================
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Keras tutorial - the Happy House\n",
"\n",
"Welcome to the first assignment of week 2. In this assignment, you will:\n",
"1. Learn to use Keras, a high-level neural networks API (programming framework), written in Python and capable of running on top of several lower-level frameworks including TensorFlow and CNTK. \n",
"2. See how you can in a couple of hours build a deep learning algorithm.\n",
"\n",
"Why are we using Keras? Keras was developed to enable deep learning engineers to build and experiment with different models very quickly. Just as TensorFlow is a higher-level framework than Python, Keras is an even higher-level framework and provides additional abstractions. Being able to go from idea to result with the least possible delay is key to finding good models. However, Keras is more restrictive than the lower-level frameworks, so there are some very complex models that you can implement in TensorFlow but not (without more difficulty) in Keras. That being said, Keras will work fine for many common models. \n",
"\n",
"In this exercise, you'll work on the \"Happy House\" problem, which we'll explain below. Let's load the required packages and solve the problem of the Happy House!"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Using TensorFlow backend.\n"
]
}
],
"source": [
"import numpy as np\n",
"from keras import layers\n",
"from keras.layers import Input, Dense, Activation, ZeroPadding2D, BatchNormalization, Flatten, Conv2D\n",
"from keras.layers import AveragePooling2D, MaxPooling2D, Dropout, GlobalMaxPooling2D, GlobalAveragePooling2D\n",
"from keras.models import Model\n",
"from keras.preprocessing import image\n",
"from keras.utils import layer_utils\n",
"from keras.utils.data_utils import get_file\n",
"from keras.applications.imagenet_utils import preprocess_input\n",
"import pydot\n",
"from IPython.display import SVG\n",
"from keras.utils.vis_utils import model_to_dot\n",
"from keras.utils import plot_model\n",
"from kt_utils import *\n",
"\n",
"import keras.backend as K\n",
"K.set_image_data_format('channels_last')\n",
"import matplotlib.pyplot as plt\n",
"from matplotlib.pyplot import imshow\n",
"\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Note**: As you can see, we've imported a lot of functions from Keras. You can use them easily just by calling them directly in the notebook. Ex: `X = Input(...)` or `X = ZeroPadding2D(...)`."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 1 - The Happy House \n",
"\n",
"For your next vacation, you decided to spend a week with five of your friends from school. It is a very convenient house with many things to do nearby. But the most important benefit is that everybody has commited to be happy when they are in the house. So anyone wanting to enter the house must prove their current state of happiness.\n",
"\n",
"<img src=\"images/happy-house.jpg\" style=\"width:350px;height:270px;\">\n",
"<caption><center> <u> <font color='purple'> **Figure 1** </u><font color='purple'> : **the Happy House**</center></caption>\n",
"\n",
"\n",
"As a deep learning expert, to make sure the \"Happy\" rule is strictly applied, you are going to build an algorithm which that uses pictures from the front door camera to check if the person is happy or not. The door should open only if the person is happy. \n",
"\n",
"You have gathered pictures of your friends and yourself, taken by the front-door camera. The dataset is labbeled. \n",
"\n",
"<img src=\"images/house-members.png\" style=\"width:550px;height:250px;\">\n",
"\n",
"Run the following code to normalize the dataset and learn about its shapes."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"number of training examples = 600\n",
"number of test examples = 150\n",
"X_train shape: (600, 64, 64, 3)\n",
"Y_train shape: (600, 1)\n",
"X_test shape: (150, 64, 64, 3)\n",
"Y_test shape: (150, 1)\n"
]
}
],
"source": [
"X_train_orig, Y_train_orig, X_test_orig, Y_test_orig, classes = load_dataset()\n",
"\n",
"# Normalize image vectors\n",
"X_train = X_train_orig/255.\n",
"X_test = X_test_orig/255.\n",
"\n",
"# Reshape\n",
"Y_train = Y_train_orig.T\n",
"Y_test = Y_test_orig.T\n",
"\n",
"print (\"number of training examples = \" + str(X_train.shape[0]))\n",
"print (\"number of test examples = \" + str(X_test.shape[0]))\n",
"print (\"X_train shape: \" + str(X_train.shape))\n",
"print (\"Y_train shape: \" + str(Y_train.shape))\n",
"print (\"X_test shape: \" + str(X_test.shape))\n",
"print (\"Y_test shape: \" + str(Y_test.shape))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Details of the \"Happy\" dataset**:\n",
"- Images are of shape (64,64,3)\n",
"- Training: 600 pictures\n",
"- Test: 150 pictures\n",
"\n",
"It is now time to solve the \"Happy\" Challenge."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 2 - Building a model in Keras\n",
"\n",
"Keras is very good for rapid prototyping. In just a short time you will be able to build a model that achieves outstanding results.\n",
"\n",
"Here is an example of a model in Keras:\n",
"\n",
"```python\n",
"def model(input_shape):\n",
" # Define the input placeholder as a tensor with shape input_shape. Think of this as your input image!\n",
" X_input = Input(input_shape)\n",
"\n",
" # Zero-Padding: pads the border of X_input with zeroes\n",
" X = ZeroPadding2D((3, 3))(X_input)\n",
"\n",
" # CONV -> BN -> RELU Block applied to X\n",
" X = Conv2D(32, (7, 7), strides = (1, 1), name = 'conv0')(X)\n",
" X = BatchNormalization(axis = 3, name = 'bn0')(X)\n",
" X = Activation('relu')(X)\n",
"\n",
" # MAXPOOL\n",
" X = MaxPooling2D((2, 2), name='max_pool')(X)\n",
"\n",
" # FLATTEN X (means convert it to a vector) + FULLYCONNECTED\n",
" X = Flatten()(X)\n",
" X = Dense(1, activation='sigmoid', name='fc')(X)\n",
"\n",
" # Create model. This creates your Keras model instance, you'll use this instance to train/test the model.\n",
" model = Model(inputs = X_input, outputs = X, name='HappyModel')\n",
" \n",
" return model\n",
"```\n",
"\n",
"Note that Keras uses a different convention with variable names than we've previously used with numpy and TensorFlow. In particular, rather than creating and assigning a new variable on each step of forward propagation such as `X`, `Z1`, `A1`, `Z2`, `A2`, etc. for the computations for the different layers, in Keras code each line above just reassigns `X` to a new value using `X = ...`. In other words, during each step of forward propagation, we are just writing the latest value in the commputation into the same variable `X`. The only exception was `X_input`, which we kept separate and did not overwrite, since we needed it at the end to create the Keras model instance (`model = Model(inputs = X_input, ...)` above). \n",
"\n",
"**Exercise**: Implement a `HappyModel()`. This assignment is more open-ended than most. We suggest that you start by implementing a model using the architecture we suggest, and run through the rest of this assignment using that as your initial model. But after that, come back and take initiative to try out other model architectures. For example, you might take inspiration from the model above, but then vary the network architecture and hyperparameters however you wish. You can also use other functions such as `AveragePooling2D()`, `GlobalMaxPooling2D()`, `Dropout()`. \n",
"\n",
"**Note**: You have to be careful with your data's shapes. Use what you've learned in the videos to make sure your convolutional, pooling and fully-connected layers are adapted to the volumes you're applying it to."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# GRADED FUNCTION: HappyModel\n",
"\n",
"def HappyModel(input_shape):\n",
" \"\"\"\n",
" Implementation of the HappyModel.\n",
" \n",
" Arguments:\n",
" input_shape -- shape of the images of the dataset\n",
"\n",
" Returns:\n",
" model -- a Model() instance in Keras\n",
" \"\"\"\n",
" \n",
" ### START CODE HERE ###\n",
" # Feel free to use the suggested outline in the text above to get started, and run through the whole\n",
" # exercise (including the later portions of this notebook) once. The come back also try out other\n",
" # network architectures as well. \n",
" # Define the input placeholder as a tensor with shape input_shape. Think of this as your input image!\n",
" X_input = Input(input_shape)\n",
"\n",
" # Zero-Padding: pads the border of X_input with zeroes\n",
" X = ZeroPadding2D((3, 3))(X_input)\n",
"\n",
" # CONV -> BN -> RELU Block applied to X\n",
" X = Conv2D(32, (7, 7), strides=(1, 1), name='conv0')(X)\n",
" X = BatchNormalization(axis=3, name='bn0')(X)\n",
" X = Activation('relu')(X)\n",
"\n",
" # MAXPOOL\n",
" X = MaxPooling2D((2, 2), name='max_pool')(X)\n",
"\n",
" # FLATTEN X (means convert it to a vector) + FULLYCONNECTED\n",
" X = Flatten()(X)\n",
" X = Dense(1, activation='sigmoid', name='fc')(X)\n",
"\n",
" # Create model. This creates your Keras model instance, you'll use this instance to train/test the model.\n",
" model = Model(inputs=X_input, outputs=X, name='HappyModel')\n",
"\n",
" return model\n",
" ### END CODE HERE ###\n",
" \n",
" return model"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"You have now built a function to describe your model. To train and test this model, there are four steps in Keras:\n",
"1. Create the model by calling the function above\n",
"2. Compile the model by calling `model.compile(optimizer = \"...\", loss = \"...\", metrics = [\"accuracy\"])`\n",
"3. Train the model on train data by calling `model.fit(x = ..., y = ..., epochs = ..., batch_size = ...)`\n",
"4. Test the model on test data by calling `model.evaluate(x = ..., y = ...)`\n",
"\n",
"If you want to know more about `model.compile()`, `model.fit()`, `model.evaluate()` and their arguments, refer to the official [Keras documentation](https://keras.io/models/model/).\n",
"\n",
"**Exercise**: Implement step 1, i.e. create the model."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"### START CODE HERE ### (1 line)\n",
"happyModel = HappyModel(X_train.shape[1:])\n",
"### END CODE HERE ###"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Exercise**: Implement step 2, i.e. compile the model to configure the learning process. Choose the 3 arguments of `compile()` wisely. Hint: the Happy Challenge is a binary classification problem."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"### START CODE HERE ### (1 line)\n",
"happyModel.compile('adam', 'binary_crossentropy', metrics=['accuracy'])\n",
"### END CODE HERE ###"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Exercise**: Implement step 3, i.e. train the model. Choose the number of epochs and the batch size."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/40\n",
"600/600 [==============================] - 12s - loss: 1.5360 - acc: 0.5900 \n",
"Epoch 2/40\n",
"600/600 [==============================] - 13s - loss: 0.4340 - acc: 0.8083 \n",
"Epoch 3/40\n",
"600/600 [==============================] - 12s - loss: 0.1771 - acc: 0.9250 \n",
"Epoch 4/40\n",
"600/600 [==============================] - 14s - loss: 0.1184 - acc: 0.9600 \n",
"Epoch 5/40\n",
"600/600 [==============================] - 14s - loss: 0.1027 - acc: 0.9617 \n",
"Epoch 6/40\n",
"600/600 [==============================] - 15s - loss: 0.0936 - acc: 0.9667 \n",
"Epoch 7/40\n",
"600/600 [==============================] - 15s - loss: 0.0744 - acc: 0.9783 \n",
"Epoch 8/40\n",
"600/600 [==============================] - 16s - loss: 0.0641 - acc: 0.9867 \n",
"Epoch 9/40\n",
"600/600 [==============================] - 14s - loss: 0.0753 - acc: 0.9733 \n",
"Epoch 10/40\n",
"600/600 [==============================] - 14s - loss: 0.0612 - acc: 0.9800 \n",
"Epoch 11/40\n",
"600/600 [==============================] - 15s - loss: 0.0519 - acc: 0.9833 \n",
"Epoch 12/40\n",
"600/600 [==============================] - 15s - loss: 0.0496 - acc: 0.9817 \n",
"Epoch 13/40\n",
"600/600 [==============================] - 15s - loss: 0.0457 - acc: 0.9900 \n",
"Epoch 14/40\n",
"600/600 [==============================] - 14s - loss: 0.0483 - acc: 0.9900 \n",
"Epoch 15/40\n",
"600/600 [==============================] - 15s - loss: 0.0329 - acc: 0.9933 \n",
"Epoch 16/40\n",
"600/600 [==============================] - 14s - loss: 0.0335 - acc: 0.9917 \n",
"Epoch 17/40\n",
"600/600 [==============================] - 15s - loss: 0.0344 - acc: 0.9867 \n",
"Epoch 18/40\n",
"600/600 [==============================] - 15s - loss: 0.0423 - acc: 0.9883 \n",
"Epoch 19/40\n",
"600/600 [==============================] - 15s - loss: 0.0282 - acc: 0.9900 \n",
"Epoch 20/40\n",
"600/600 [==============================] - 15s - loss: 0.0232 - acc: 0.9933 \n",
"Epoch 21/40\n",
"600/600 [==============================] - 15s - loss: 0.0206 - acc: 0.9967 \n",
"Epoch 22/40\n",
"600/600 [==============================] - 15s - loss: 0.0258 - acc: 0.9917 \n",
"Epoch 23/40\n",
"600/600 [==============================] - 15s - loss: 0.0179 - acc: 0.9950 \n",
"Epoch 24/40\n",
"600/600 [==============================] - 15s - loss: 0.0159 - acc: 0.9967 \n",
"Epoch 25/40\n",
"600/600 [==============================] - 15s - loss: 0.0206 - acc: 0.9950 \n",
"Epoch 26/40\n",
"600/600 [==============================] - 16s - loss: 0.0158 - acc: 1.0000 \n",
"Epoch 27/40\n",
"600/600 [==============================] - 15s - loss: 0.0166 - acc: 0.9917 \n",
"Epoch 28/40\n",
"600/600 [==============================] - 15s - loss: 0.0146 - acc: 0.9983 \n",
"Epoch 29/40\n",
"600/600 [==============================] - 17s - loss: 0.0206 - acc: 0.9950 \n",
"Epoch 30/40\n",
"600/600 [==============================] - 16s - loss: 0.0411 - acc: 0.9867 \n",
"Epoch 31/40\n",
"600/600 [==============================] - 15s - loss: 0.0268 - acc: 0.9933 \n",
"Epoch 32/40\n",
"600/600 [==============================] - 15s - loss: 0.0623 - acc: 0.9800 \n",
"Epoch 33/40\n",
"600/600 [==============================] - 15s - loss: 0.0984 - acc: 0.9567 \n",
"Epoch 34/40\n",
"600/600 [==============================] - 15s - loss: 0.0363 - acc: 0.9933 \n",
"Epoch 35/40\n",
"600/600 [==============================] - 16s - loss: 0.0343 - acc: 0.9883 \n",
"Epoch 36/40\n",
"600/600 [==============================] - 17s - loss: 0.0424 - acc: 0.9867 \n",
"Epoch 37/40\n",
"600/600 [==============================] - 18s - loss: 0.0277 - acc: 0.9933 \n",
"Epoch 38/40\n",
"600/600 [==============================] - 18s - loss: 0.0104 - acc: 1.0000 \n",
"Epoch 39/40\n",
"600/600 [==============================] - 15s - loss: 0.0152 - acc: 0.9950 \n",
"Epoch 40/40\n",
"600/600 [==============================] - 15s - loss: 0.0161 - acc: 0.9917 \n"
]
},
{
"data": {
"text/plain": [
"<keras.callbacks.History at 0x7fda947bed68>"
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"### START CODE HERE ### (1 line)\n",
"happyModel.fit(X_train, Y_train, epochs=40, batch_size=50)\n",
"### END CODE HERE ###"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Note that if you run `fit()` again, the `model` will continue to train with the parameters it has already learnt instead of reinitializing them.\n",
"\n",
"**Exercise**: Implement step 4, i.e. test/evaluate the model."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"scrolled": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"150/150 [==============================] - 2s \n",
"\n",
"Loss = 0.200629468759\n",
"Test Accuracy = 0.92666667064\n"
]
}
],
"source": [
"### START CODE HERE ### (1 line)\n",
"preds = happyModel.evaluate(X_test, Y_test, batch_size=32, verbose=1, sample_weight=None)\n",
"### END CODE HERE ###\n",
"print()\n",
"print (\"Loss = \" + str(preds[0]))\n",
"print (\"Test Accuracy = \" + str(preds[1]))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If your `happyModel()` function worked, you should have observed much better than random-guessing (50%) accuracy on the train and test sets. To pass this assignment, you have to get at least 75% accuracy. \n",
"\n",
"To give you a point of comparison, our model gets around **95% test accuracy in 40 epochs** (and 99% train accuracy) with a mini batch size of 16 and \"adam\" optimizer. But our model gets decent accuracy after just 2-5 epochs, so if you're comparing different models you can also train a variety of models on just a few epochs and see how they compare. \n",
"\n",
"If you have not yet achieved 75% accuracy, here're some things you can play around with to try to achieve it:\n",
"\n",
"- Try using blocks of CONV->BATCHNORM->RELU such as:\n",
"```python\n",
"X = Conv2D(32, (3, 3), strides = (1, 1), name = 'conv0')(X)\n",
"X = BatchNormalization(axis = 3, name = 'bn0')(X)\n",
"X = Activation('relu')(X)\n",
"```\n",
"until your height and width dimensions are quite low and your number of channels quite large (≈32 for example). You are encoding useful information in a volume with a lot of channels. You can then flatten the volume and use a fully-connected layer.\n",
"- You can use MAXPOOL after such blocks. It will help you lower the dimension in height and width.\n",
"- Change your optimizer. We find Adam works well. \n",
"- If the model is struggling to run and you get memory issues, lower your batch_size (12 is usually a good compromise)\n",
"- Run on more epochs, until you see the train accuracy plateauing. \n",
"\n",
"Even if you have achieved 75% accuracy, please feel free to keep playing with your model to try to get even better results. \n",
"\n",
"**Note**: If you perform hyperparameter tuning on your model, the test set actually becomes a dev set, and your model might end up overfitting to the test (dev) set. But just for the purpose of this assignment, we won't worry about that here.\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 3 - Conclusion\n",
"\n",
"Congratulations, you have solved the Happy House challenge! \n",
"\n",
"Now, you just need to link this model to the front-door camera of your house. We unfortunately won't go into the details of how to do that here. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"<font color='blue'>\n",
"**What we would like you to remember from this assignment:**\n",
"- Keras is a tool we recommend for rapid prototyping. It allows you to quickly try out different model architectures. Are there any applications of deep learning to your daily life that you'd like to implement using Keras? \n",
"- Remember how to code a model in Keras and the four steps leading to the evaluation of your model on the test set. Create->Compile->Fit/Train->Evaluate/Test."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 4 - Test with your own image (Optional)\n",
"\n",
"Congratulations on finishing this assignment. You can now take a picture of your face and see if you could enter the Happy House. To do that:\n",
" 1. Click on \"File\" in the upper bar of this notebook, then click \"Open\" to go on your Coursera Hub.\n",
" 2. Add your image to this Jupyter Notebook's directory, in the \"images\" folder\n",
" 3. Write your image's name in the following code\n",
" 4. Run the code and check if the algorithm is right (0 is unhappy, 1 is happy)!\n",
" \n",
"The training/test sets were quite similar; for example, all the pictures were taken against the same background (since a front door camera is always mounted in the same position). This makes the problem easier, but a model trained on this data may or may not work on your own data. But feel free to give it a try! "
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[[ 0.]]\n"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAP8AAAD8CAYAAAC4nHJkAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nF7+GRori/wOMhrX8XuEH+AAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x7fda941ec160>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"### START CODE HERE ###\n",
"img_path = 'images/my_image.jpg'\n",
"### END CODE HERE ###\n",
"img = image.load_img(img_path, target_size=(64, 64))\n",
"imshow(img)\n",
"\n",
"x = image.img_to_array(img)\n",
"x = np.expand_dims(x, axis=0)\n",
"x = preprocess_input(x)\n",
"\n",
"print(happyModel.predict(x))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## 5 - Other useful functions in Keras (Optional)\n",
"\n",
"Two other basic features of Keras that you'll find useful are:\n",
"- `model.summary()`: prints the details of your layers in a table with the sizes of its inputs/outputs\n",
"- `plot_model()`: plots your graph in a nice layout. You can even save it as \".png\" using SVG() if you'd like to share it on social media ;). It is saved in \"File\" then \"Open...\" in the upper bar of the notebook.\n",
"\n",
"Run the following code."
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"input_1 (InputLayer) (None, 64, 64, 3) 0 \n",
"_________________________________________________________________\n",
"zero_padding2d_1 (ZeroPaddin (None, 70, 70, 3) 0 \n",
"_________________________________________________________________\n",
"conv0 (Conv2D) (None, 64, 64, 32) 4736 \n",
"_________________________________________________________________\n",
"bn0 (BatchNormalization) (None, 64, 64, 32) 128 \n",
"_________________________________________________________________\n",
"activation_1 (Activation) (None, 64, 64, 32) 0 \n",
"_________________________________________________________________\n",
"max_pool (MaxPooling2D) (None, 32, 32, 32) 0 \n",
"_________________________________________________________________\n",
"flatten_1 (Flatten) (None, 32768) 0 \n",
"_________________________________________________________________\n",
"fc (Dense) (None, 1) 32769 \n",
"=================================================================\n",
"Total params: 37,633\n",
"Trainable params: 37,569\n",
"Non-trainable params: 64\n",
"_________________
gitextract_stf4uzcu/
├── .gitignore
├── Convolutional Neural Networks/
│ ├── Convolution model - Application - v1.ipynb
│ ├── Convolution model - Step by Step - v1.ipynb
│ ├── Keras - Tutorial - Happy House v1.ipynb
│ └── Residual Networks - v1.ipynb
├── Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/
│ ├── Gradient Checking.ipynb
│ ├── Initialization.ipynb
│ ├── Optimization methods.ipynb
│ ├── Regularization.ipynb
│ ├── Tensorflow Tutorial.ipynb
│ ├── Week 1 Quiz - Practical aspects of deep learning.md
│ ├── Week 2 Quiz - Optimization algorithms.md
│ └── Week 3 Quiz - Hyperparameter tuning, Batch Normalization, Programming Frameworks.md
├── LICENSE
├── Neural Networks and Deep Learning/
│ ├── Building your Deep Neural Network - Step by Step.ipynb
│ ├── Deep Neural Network - Application.ipynb
│ ├── Logistic Regression with a Neural Network mindset.ipynb
│ ├── Planar data classification with one hidden layer.ipynb
│ ├── Week 1 Quiz - Introduction to deep learning.md
│ ├── Week 2 Quiz - Neural Network Basics.md
│ ├── Week 3 Quiz - Shallow Neural Networks.md
│ └── Week 4 Quiz - Key concepts on Deep Neural Networks.md
├── README.md
├── Sequence Models/
│ ├── Building a Recurrent Neural Network - Step by Step - v2.ipynb
│ ├── Dinosaurus Island -- Character level language model final - v3.ipynb
│ ├── Emojify - v2.ipynb
│ ├── Improvise a Jazz Solo with an LSTM Network - v1.ipynb
│ ├── Neural machine translation with attention - v2.ipynb
│ ├── Operations on word vectors - v2.ipynb
│ ├── Trigger word detection - v1.ipynb
│ └── rnn_utils.py
└── Structuring Machine Learning Projects/
├── Week 1 Quiz - Bird recognition in the city of Peacetopia (case study).md
└── Week 2 Quiz - Autonomous driving (case study).md
SYMBOL INDEX (4 symbols across 1 files) FILE: Sequence Models/rnn_utils.py function softmax (line 3) | def softmax(x): function sigmoid (line 8) | def sigmoid(x): function initialize_adam (line 12) | def initialize_adam(parameters) : function update_parameters_with_adam (line 49) | def update_parameters_with_adam(parameters, grads, v, s, t, learning_rat...
Condensed preview — 33 files, each showing path, character count, and a content snippet. Download the .json file or copy for the full structured content (8,014K chars).
[
{
"path": ".gitignore",
"chars": 1174,
"preview": "# Byte-compiled / optimized / DLL files\n__pycache__/\n*.py[cod]\n*$py.class\n\n# C extensions\n*.so\n\n.idea\n.DS_Store\n\n# Distr"
},
{
"path": "Convolutional Neural Networks/Convolution model - Application - v1.ipynb",
"chars": 87054,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Convolutional Neural Networks: Ap"
},
{
"path": "Convolutional Neural Networks/Convolution model - Step by Step - v1.ipynb",
"chars": 57767,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Convolutional Neural Networks: St"
},
{
"path": "Convolutional Neural Networks/Keras - Tutorial - Happy House v1.ipynb",
"chars": 60786,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Keras tutorial - the Happy House\\"
},
{
"path": "Convolutional Neural Networks/Residual Networks - v1.ipynb",
"chars": 347018,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Residual Networks\\n\",\n \"\\n\",\n "
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Gradient Checking.ipynb",
"chars": 26891,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Gradient Checking\\n\",\n \"\\n\",\n "
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Initialization.ipynb",
"chars": 274786,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Initialization\\n\",\n \"\\n\",\n "
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Optimization methods.ipynb",
"chars": 383779,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Optimization Methods\\n\",\n \"\\n\""
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Regularization.ipynb",
"chars": 273110,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Regularization\\n\",\n \"\\n\",\n "
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Tensorflow Tutorial.ipynb",
"chars": 210764,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# TensorFlow Tutorial\\n\",\n \"\\n\","
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Week 1 Quiz - Practical aspects of deep learning.md",
"chars": 2111,
"preview": "## Week 1 Quiz - Practical aspects of deep learning\n\n1. If you have 10,000,000 examples, how would you split the train/d"
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Week 2 Quiz - Optimization algorithms.md",
"chars": 3989,
"preview": "## Week 2 Quiz - Optimization algorithms\n\n1. Which notation would you use to denote the 3rd layer’s activations when the"
},
{
"path": "Improving Deep Neural Networks Hyperparameter tuning, Regularization and Optimization/Week 3 Quiz - Hyperparameter tuning, Batch Normalization, Programming Frameworks.md",
"chars": 3475,
"preview": "## Week 3 Quiz - Hyperparameter tuning, Batch Normalization, Programming Frameworks\n\n1. If searching among a large numbe"
},
{
"path": "LICENSE",
"chars": 1064,
"preview": "MIT License\n\nCopyright (c) 2017 Ji Yang\n\nPermission is hereby granted, free of charge, to any person obtaining a copy\nof"
},
{
"path": "Neural Networks and Deep Learning/Building your Deep Neural Network - Step by Step.ipynb",
"chars": 57213,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Building your Deep Neural Network"
},
{
"path": "Neural Networks and Deep Learning/Deep Neural Network - Application.ipynb",
"chars": 2009547,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Deep Neural Network for Image Cla"
},
{
"path": "Neural Networks and Deep Learning/Logistic Regression with a Neural Network mindset.ipynb",
"chars": 784897,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Logistic Regression with a Neural"
},
{
"path": "Neural Networks and Deep Learning/Planar data classification with one hidden layer.ipynb",
"chars": 1074428,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Planar data classification with o"
},
{
"path": "Neural Networks and Deep Learning/Week 1 Quiz - Introduction to deep learning.md",
"chars": 4007,
"preview": "## Week 1 Quiz - Introduction to deep learning\n\n1. What does the analogy “AI is the new electricity” refer to?\n\n - [ "
},
{
"path": "Neural Networks and Deep Learning/Week 2 Quiz - Neural Network Basics.md",
"chars": 3300,
"preview": "## Week 2 Quiz - Neural Network Basics\n\n1. What does a neuron compute?\n\n - [ ] A neuron computes an activation functi"
},
{
"path": "Neural Networks and Deep Learning/Week 3 Quiz - Shallow Neural Networks.md",
"chars": 5752,
"preview": "## Week 3 Quiz - Shallow Neural Networks\n\n1. Which of the following are true? (Check all that apply.) **Notice that I o"
},
{
"path": "Neural Networks and Deep Learning/Week 4 Quiz - Key concepts on Deep Neural Networks.md",
"chars": 4752,
"preview": "## Week 4 Quiz - Key concepts on Deep Neural Networks\n\n1. What is the \"cache\" used for in our implementation of forward "
},
{
"path": "README.md",
"chars": 8880,
"preview": "# Deep Learning Specialization on Coursera\n\n**Master Deep Learning, and Break into AI**\n\nInstructor: [Andrew Ng](http://"
},
{
"path": "Sequence Models/Building a Recurrent Neural Network - Step by Step - v2.ipynb",
"chars": 86304,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Building your Recurrent Neural Ne"
},
{
"path": "Sequence Models/Dinosaurus Island -- Character level language model final - v3.ipynb",
"chars": 46528,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Character level language model - "
},
{
"path": "Sequence Models/Emojify - v2.ipynb",
"chars": 69210,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Emojify! \\n\",\n \"\\n\",\n \"Welc"
},
{
"path": "Sequence Models/Improvise a Jazz Solo with an LSTM Network - v1.ipynb",
"chars": 1798293,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Improvise a Jazz Solo with an LST"
},
{
"path": "Sequence Models/Neural machine translation with attention - v2.ipynb",
"chars": 90389,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Neural Machine Translation\\n\",\n "
},
{
"path": "Sequence Models/Operations on word vectors - v2.ipynb",
"chars": 30267,
"preview": "{\n \"cells\": [\n {\n \"cell_type\": \"markdown\",\n \"metadata\": {},\n \"source\": [\n \"# Operations on word vectors\\n\",\n "
},
{
"path": "Sequence Models/rnn_utils.py",
"chars": 5155,
"preview": "import numpy as np\n\ndef softmax(x):\n e_x = np.exp(x - np.max(x))\n return e_x / e_x.sum(axis=0)\n\n\ndef sigmoid(x):\n "
},
{
"path": "Structuring Machine Learning Projects/Week 1 Quiz - Bird recognition in the city of Peacetopia (case study).md",
"chars": 6786,
"preview": "## Week 1 Quiz - Bird recognition in the city of Peacetopia (case study)\n\n1. Having three evaluation metrics makes it ha"
},
{
"path": "Structuring Machine Learning Projects/Week 2 Quiz - Autonomous driving (case study).md",
"chars": 11526,
"preview": "## Week 2 Quiz - Autonomous driving (case study)\n\n1. You are just getting started on this project. What is the first thi"
}
]
// ... and 1 more files (download for full content)
About this extraction
This page contains the full source code of the Kulbear/deep-learning-coursera GitHub repository, extracted and formatted as plain text for AI agents and large language models (LLMs). The extraction includes 33 files (28.3 MB), approximately 2.0M tokens, and a symbol index with 4 extracted functions, classes, methods, constants, and types. Use this with OpenClaw, Claude, ChatGPT, Cursor, Windsurf, or any other AI tool that accepts text input. You can copy the full output to your clipboard or download it as a .txt file.
Extracted by GitExtract — free GitHub repo to text converter for AI. Built by Nikandr Surkov.