In this notebook, a template is provided for you to implement your functionality in stages which is required to successfully complete this project. If additional code is required that cannot be included in the notebook, be sure that the Python code is successfully imported and included in your submission, if necessary. Sections that begin with 'Implementation' in the header indicate where you should begin your implementation for your project. Note that some sections of implementation are optional, and will be marked with 'Optional' in the header.
In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.
Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. In addition, Markdown cells can be edited by typically double-clicking the cell to enter edit mode.
In [1]:
# Load pickled data
import pickle
import csv
# TODO: Fill this in based on where you saved the training and testing data
training_file = 'train.p'
testing_file = 'test.p'
label_text = {}
with open('signnames.csv') as csvfile:
reader = csv.DictReader(csvfile)
for row in reader:
label_text[row['ClassId']]=row['SignName']
with open(training_file, mode='rb') as f:
train = pickle.load(f)
with open(testing_file, mode='rb') as f:
test = pickle.load(f)
X_train, y_train = train['features'], train['labels']
X_test, y_test = test['features'], test['labels']
The pickled data is a dictionary with 4 key/value pairs:
'features' is a 4D array containing raw pixel data of the traffic sign images, (num examples, width, height, channels).'labels' is a 1D array containing the label/class id of the traffic sign. The file signnames.csv contains id -> name mappings for each id.'sizes' is a list containing tuples, (width, height) representing the the original width and height the image.'coords' is a list containing tuples, (x1, y1, x2, y2) representing coordinates of a bounding box around the sign in the image. THESE COORDINATES ASSUME THE ORIGINAL IMAGE. THE PICKLED DATA CONTAINS RESIZED VERSIONS (32 by 32) OF THESE IMAGESComplete the basic data summary below.
In [2]:
### Replace each question mark with the appropriate value.
import numpy as np
# TODO: Number of training examples
n_train = X_train.shape[0]
# TODO: Number of testing examples.
n_test = X_test.shape[0]
# TODO: What's the shape of an traffic sign image?
image_shape = X_train[0].shape
# TODO: How many unique classes/labels there are in the dataset.
n_classes = len(np.unique(y_train))
print("Number of training examples =", n_train)
print("Number of testing examples =", n_test)
print("Image data shape =", image_shape)
print("Number of classes =", n_classes)
Visualize the German Traffic Signs Dataset using the pickled file(s). This is open ended, suggestions include: plotting traffic sign images, plotting the count of each sign, etc.
The Matplotlib examples and gallery pages are a great resource for doing visualizations in Python.
NOTE: It's recommended you start with something simple first. If you wish to do more, come back to it after you've completed the rest of the sections.
In [3]:
### Data exploration visualization goes here.
### Feel free to use as many code cells as needed.
import matplotlib.pyplot as plt
import random
# Visualizations will be shown in the notebook.
%matplotlib inline
index = random.randint(0, len(X_train))
image = X_train[index].squeeze()
plt.figure(figsize=(1,1))
plt.imshow(image)
print(y_train[index])
In [4]:
import matplotlib.pyplot as plt
plt.figure(figsize=(20,20))
for i in range(43):
plt.subplot(7,7,i+1)
ii = np.where(y_train== i)[0]
plt_image = X_train[ii[int(np.random.random()*len(ii))]].squeeze()
plt.title(label_text[str(i)])
plt.imshow(plt_image)
In [5]:
from collections import Counter
plt.bar(Counter(y_train).keys(), Counter(y_train).values())
plt.xlim(0, 43)
Out[5]:
Design and implement a deep learning model that learns to recognize traffic signs. Train and test your model on the German Traffic Sign Dataset.
There are various aspects to consider when thinking about this problem:
Here is an example of a published baseline model on this problem. It's not required to be familiar with the approach used in the paper but, it's good practice to try to read papers like these.
NOTE: The LeNet-5 implementation shown in the classroom at the end of the CNN lesson is a solid starting point. You'll have to change the number of classes and possibly the preprocessing, but aside from that it's plug and play!
In [6]:
### Preprocess the data here.
### Feel free to use as many code cells as needed.
from sklearn.model_selection import train_test_split
from sklearn.utils import shuffle
from tensorflow.examples.tutorials.mnist import input_data
import cv2
def generate_fake_sample(fake_y_value):
'''
INPUT:
X - your oversampled data
OUTPUT:
random_sample_X, fake_y_lable
'''
mnist = input_data.read_data_sets("MNIST_data/", reshape=False)
X_fake = mnist.test.images[:3000]
X_fake = np.pad(X_fake, ((0,0),(2,2),(2,2),(0,0)), 'constant')
X_fake = np.resize(X_fake, (3000,32,32,3))
return X_fake, fake_y_value*np.ones(X_fake.shape[0])
def oversample(X, y):
"""
INPUT:
X, y - your data
OUTPUT:
X,y - oversampled data
`oversample` randomly replicates observations
in X, y to achieve equal portion in each label
"""
target_count = Counter(y).most_common()[0][1]
X_resampled = X.copy()
y_resampled = y.copy()
for y_value, y_count in Counter(y).most_common():
number_of_new_observations = target_count - y_count
obs_indices = np.where(y==y_value)[0]
new_obs_indices = np.random.choice(obs_indices,
size=number_of_new_observations,
replace=True)
X_new, y_new = X[new_obs_indices], y[new_obs_indices]
X_resampled = np.vstack((X_resampled, X_new))
y_resampled = np.concatenate((y_resampled,y_new))
return X_resampled, y_resampled
def shuffle_train_data(X_t,y_t):
return shuffle(X_t,y_t)
#default sampling has similar accuracy but oversampling is a bit better.
# X_t, X_v, y_t, y_v = train_test_split(X_train, y_train, test_size = 0.2)
X_oversampled, y_oversampled = oversample(X_train, y_train)
# X_fake, y_fake = generate_fake_sample(43)
# X_resampled = np.vstack((X_oversampled, X_fake))
# y_resampled = np.concatenate((y_fake, y_oversampled))
#Fake data does not work too great in this case
#X_t, X_v, y_t, y_v = train_test_split(X_resampled, y_resampled, test_size = 0.2)
#best with oversample only so far
X_t, X_v, y_t, y_v = train_test_split(X_oversampled, y_oversampled, test_size = 0.2)
In [7]:
# print('original size: ', y_train.shape[0], 'new size: ', y_resampled.shape[0])
# plt.bar(Counter(y_resampled).keys(), Counter(y_resampled).values())
# plt.xlim(0, 44)
In [8]:
import tensorflow as tf
EPOCHS = 50
BATCH_SIZE = 128
Implement the LeNet-5 neural network architecture.
This is the only cell you need to edit.
The LeNet architecture accepts a 32x32xC image as input, where C is the number of color channels. Since images are rbg, C is 3 in this case.
Layer 1: Convolutional. The output shape should be 28x28x6.
Activation. Your choice of activation function.
Pooling. The output shape should be 14x14x6.
Layer 2: Convolutional. The output shape should be 10x10x16.
Activation. Your choice of activation function.
Pooling. The output shape should be 5x5x16.
Flatten. Flatten the output shape of the final pooling layer such that it's 1D instead of 3D. The easiest way to do is by using tf.contrib.layers.flatten, which is already imported for you.
Layer 3: Fully Connected. This should have 120 outputs.
Activation. Your choice of activation function.
Layer 4: Fully Connected. This should have 84 outputs.
Activation. Your choice of activation function.
Layer 5: Fully Connected (Logits). This should have 44 outputs.
Return the result of the 2nd fully connected layer.
In [9]:
from tensorflow.contrib.layers import flatten
mu = 0
sigma = 0.1
weights = {'wc1': tf.Variable(tf.truncated_normal(shape=(5, 5, 3, 6), mean = mu, stddev = sigma)),
'wc2': tf.Variable(tf.truncated_normal([5, 5, 6, 16],mean = mu, stddev = sigma)),
'wd1': tf.Variable(tf.truncated_normal([5*5*16, 120],mean = mu, stddev = sigma)),
'wd2': tf.Variable(tf.truncated_normal([120, 84],mean = mu, stddev = sigma)),
'out': tf.Variable(tf.truncated_normal([84, 44],mean = mu, stddev = sigma))}
biases = {
'bc1': tf.Variable(tf.zeros([6])),
'bc2': tf.Variable(tf.zeros([16])),
'bd1': tf.Variable(tf.zeros([120])),
'bd2': tf.Variable(tf.zeros([84])),
'out': tf.Variable(tf.zeros([44]))}
def conv2d(x, W, b, strides=1):
x = tf.nn.conv2d(x, W, strides=[1, strides, strides, 1], padding='VALID')
x = tf.nn.bias_add(x, b)
return tf.nn.relu(x)
def maxpool2d(x, k=2):
return tf.nn.max_pool(
x,
ksize=[1, k, k, 1],
strides=[1, k, k, 1],
padding='VALID')
def LeNet(x):
# Arguments used for tf.truncated_normal, randomly defines variables for the weights and biases for each layer
# TODO: Layer 1: Convolutional. Input = 32x32x3. Output = 28x28x6.
#new_height = (input_height - filter_height)/S + 1
conv1 = conv2d(x, weights['wc1'], biases['bc1'])
# TODO: Activation.
# TODO: Pooling. Input = 28x28x6. Output = 14x14x6.
conv1 = maxpool2d(conv1, k=2)
# TODO: Layer 2: Convolutional. Output = 10x10x16.
# TODO: Activation.
conv2 = conv2d(conv1, weights['wc2'], biases['bc2'])
# TODO: Pooling. Input = 10x10x16. Output = 5x5x16.
conv2 = maxpool2d(conv2, k=2)
# TODO: Flatten. Input = 5x5x16. Output = 400.
fc1 = flatten(conv2)
# TODO: Layer 3: Fully Connected. Input = 400. Output = 120.
fc1 = tf.add(tf.matmul(fc1, weights['wd1']), biases['bd1'])
fc1 = tf.nn.relu(fc1)
# TODO: Layer 4: Fully Connected. Input = 120. Output = 84.
fc2 = tf.add(tf.matmul(fc1, weights['wd2']), biases['bd2'])
fc2 = tf.nn.relu(fc2)
# TODO: Layer 5: Fully Connected. Input = 84. Output = 44.
logits = tf.add(tf.matmul(fc2, weights['out']), biases['out'])
return logits
In [10]:
# from tensorflow.contrib.layers import flatten
# def LeNet(x):
# # Arguments used for tf.truncated_normal, randomly defines variables for the weights and biases for each layer
# mu = 0
# sigma = 0.1
# # SOLUTION: Layer 1: Convolutional. Input = 32x32x1. Output = 28x28x6.
# conv1_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 3, 6), mean = mu, stddev = sigma))
# conv1_b = tf.Variable(tf.zeros(6))
# conv1 = tf.nn.conv2d(x, conv1_W, strides=[1, 1, 1, 1], padding='VALID') + conv1_b
# # SOLUTION: Activation.
# conv1 = tf.nn.relu(conv1)
# # SOLUTION: Pooling. Input = 28x28x6. Output = 14x14x6.
# conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')
# # SOLUTION: Layer 2: Convolutional. Output = 10x10x16.
# conv2_W = tf.Variable(tf.truncated_normal(shape=(5, 5, 6, 16), mean = mu, stddev = sigma))
# conv2_b = tf.Variable(tf.zeros(16))
# conv2 = tf.nn.conv2d(conv1, conv2_W, strides=[1, 1, 1, 1], padding='VALID') + conv2_b
# # SOLUTION: Activation.
# conv2 = tf.nn.relu(conv2)
# # SOLUTION: Pooling. Input = 10x10x16. Output = 5x5x16.
# conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID')
# # SOLUTION: Flatten. Input = 5x5x16. Output = 400.
# fc0 = flatten(conv2)
# # SOLUTION: Layer 3: Fully Connected. Input = 400. Output = 120.
# fc1_W = tf.Variable(tf.truncated_normal(shape=(400, 120), mean = mu, stddev = sigma))
# fc1_b = tf.Variable(tf.zeros(120))
# fc1 = tf.matmul(fc0, fc1_W) + fc1_b
# # SOLUTION: Activation.
# fc1 = tf.nn.relu(fc1)
# # SOLUTION: Layer 4: Fully Connected. Input = 120. Output = 84.
# fc2_W = tf.Variable(tf.truncated_normal(shape=(120, 84), mean = mu, stddev = sigma))
# fc2_b = tf.Variable(tf.zeros(84))
# fc2 = tf.matmul(fc1, fc2_W) + fc2_b
# # SOLUTION: Activation.
# fc2 = tf.nn.relu(fc2)
# # SOLUTION: Layer 5: Fully Connected. Input = 84. Output = 10.
# fc3_W = tf.Variable(tf.truncated_normal(shape=(84, 44), mean = mu, stddev = sigma))
# fc3_b = tf.Variable(tf.zeros(44))
# logits = tf.matmul(fc2, fc3_W) + fc3_b
# return logits
In [11]:
x = tf.placeholder(tf.float32, (None, 32, 32, 3))
y = tf.placeholder(tf.int32, (None))
one_hot_y = tf.one_hot(y, 44)
# keep_prob = tf.placeholder(tf.float32)
In [12]:
rate = 0.001
logits = LeNet(x)
cross_entropy = tf.nn.softmax_cross_entropy_with_logits(logits, one_hot_y)
loss_operation = tf.reduce_mean(cross_entropy)
optimizer = tf.train.AdamOptimizer(learning_rate = rate)
training_operation = optimizer.minimize(loss_operation)
In [13]:
correct_prediction = tf.equal(tf.argmax(logits, 1), tf.argmax(one_hot_y, 1))
accuracy_operation = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
saver = tf.train.Saver()
def evaluate(X_data, y_data):
num_examples = len(X_data)
total_accuracy = 0
sess = tf.get_default_session()
for offset in range(0, num_examples, BATCH_SIZE):
batch_x, batch_y = X_data[offset:offset+BATCH_SIZE], y_data[offset:offset+BATCH_SIZE]
accuracy = sess.run(accuracy_operation, feed_dict={x: batch_x, y: batch_y})
total_accuracy += (accuracy * len(batch_x))
return total_accuracy / num_examples
In [14]:
import time
start = time.clock()
with tf.Session() as sess:
sess.run(tf.initialize_all_variables())
num_examples = len(X_t)
print("Training...")
print()
for i in range(EPOCHS):
X_t, y_t = shuffle(X_t, y_t)
for offset in range(0, num_examples, BATCH_SIZE):
end = offset + BATCH_SIZE
batch_x, batch_y = X_t[offset:end], y_t[offset:end]
sess.run(training_operation, feed_dict={x: batch_x, y: batch_y})
validation_accuracy = evaluate(X_v, y_v)
print("EPOCH {} ...".format(i+1))
print("Validation Accuracy = {:.3f}".format(validation_accuracy))
print()
saver.save(sess, './lenet50')
print("Model saved")
end = time.clock()
print ("Time used: {:2f}s".format(end - start))
In [15]:
with tf.Session() as sess:
saver.restore(sess, 'lenet50')
test_accuracy = evaluate(X_test, y_test)
print("Test Accuracy = {:.3f}".format(test_accuracy))
Answer:
In [16]:
### Generate additional data (OPTIONAL!)
### and split the data into training/validation/testing sets here.
### Feel free to use as many code cells as needed.
I loaded and confirmed data shape, and plotted images for visualization. I used over_sampling to generate additional data to make the count of each label equally. In addition, I created fake data to prevent overfitting. I chose train_test_split to split the training data to 80% train data and 20% valid data.
Answer:
In [17]:
### Define your architecture here.
### Feel free to use as many code cells as needed.
I genreated additional data with oversampling and fake data with random sampling approx. 6000 data point. These random data can help to prevent overfitting. I make sure each data label have the same count number as the most common label. The new dataset is bigger than the orignal one since I ovsersampled the undersize data. I used sklearn library for train_split test to create
What does your final architecture look like? (Type of model, layers, sizes, connectivity, etc.) For reference on how to build a deep neural network using TensorFlow, see Deep Neural Network in TensorFlow from the classroom.
Answer:
In [18]:
### Train your model here.
### Feel free to use as many code cells as needed.
I use lenet as my model. It has 5 layers.
The first layer, it takes input of a image with 32x32x3, run the convolution the image to output a image with 28x28x6 and apply relu function. after that, I do pooling to make the image become 14x14x6.
The second layer is doing the same thing, it runs convolution of the image 10x10x16 and apply relu function, then do pooling again and output as 5x5x16.
The third layer, I connect the image together and apply relu and liner equation. The fourth and final layer are doing the same thing.
Answer: EPOCHS = 150, BATCH_SIZE = 128, rate = 0.001, optimizer = AdamOptimizer, loss_operation = soft_max_entropy_with_logits
What approach did you take in coming up with a solution to this problem? It may have been a process of trial and error, in which case, outline the steps you took to get to the final solution and why you chose those steps. Perhaps your solution involved an already well known implementation or architecture. In this case, discuss why you think this is suitable for the current problem.
Answer:
LeNet is great implementation for trainning 32x32 size images. I just replaced the imput layer and output layer size to match the need for this traffic sign classfiication. I have tried different layer size and tuned the parameters with the best validated accuracy.
Take several pictures of traffic signs that you find on the web or around you (at least five), and run them through your classifier on your computer to produce example results. The classifier might not recognize some local signs but it could prove interesting nonetheless.
You may find signnames.csv useful as it contains mappings from the class id (integer) to the actual sign name.
In [19]:
### Load the images and plot them here.
### Feel free to use as many code cells as needed.
import os
import matplotlib.image as mpimg
plt.figure(figsize=(9,9))
label_for_new_image = [27,40,17,2,31,23,35,25,24,4]
file = os.listdir("web_images/")
file = [x for x in file if not x.startswith(".")]
new_image_together = []
for i , pic in enumerate(file):
image = cv2.imread("web_images/"+pic)
resized_image = cv2.resize(image, (32, 32))
resized_image = cv2.cvtColor(resized_image, cv2.COLOR_RGB2BGR)
plt.subplot(2,5,i+1)
plt.title(label_text[str(label_for_new_image[i])])
plt.imshow(resized_image)
new_image_together.append(resized_image)
images_together_np=np.asarray(new_image_together)
print(images_together_np.shape)
# (10, 32, 32, 3)
Answer:
In [20]:
### Run the predictions here.
### Feel free to use as many code cells as needed.
I have 10 image from the web some of them are in similiar to the training set but for image 2,4,5,7 all have difference from the training set
I predict those image maybe difficult to classify like 4 and 5, because they have shape and color is different.
And for 2 and 7 both have addition information on the image, but I think those addition information will not affect the prediction.
Is your model able to perform equally well on captured pictures when compared to testing on the dataset? The simplest way to do this check the accuracy of the predictions. For example, if the model predicted 1 out of 5 signs correctly, it's 20% accurate.
NOTE: You could check the accuracy manually by using signnames.csv (same directory). This file has a mapping from the class id (0-42) to the corresponding sign name. So, you could take the class id the model outputs, lookup the name in signnames.csv and see if it matches the sign from the image.
In [ ]:
In [21]:
with tf.Session() as sess:
saver.restore(sess, "lenet50")
test_accuracy = evaluate(images_together_np, label_for_new_image)
print("Accuracy on new images = {:.3f}".format(test_accuracy))
Answer:
In [22]:
### Visualize the softmax probabilities here.
### Feel free to use as many code cells as needed.
I run the test for the accuracy for the new image which I get from the internet. I run it for 10 images and the result is only 30% correct. It is under my expectation, because my model has a high validation accuracy and test accuracy on the testing data.
Use the model's softmax probabilities to visualize the certainty of its predictions, tf.nn.top_k could prove helpful here. Which predictions is the model certain of? Uncertain? If the model was incorrect in its initial prediction, does the correct prediction appear in the top k? (k should be 5 at most)
tf.nn.top_k will return the values and indices (class ids) of the top k predictions. So if k=3, for each sign, it'll return the 3 largest probabilities (out of a possible 43) and the correspoding class ids.
Take this numpy array as an example:
# (5, 6) array
a = np.array([[ 0.24879643, 0.07032244, 0.12641572, 0.34763842, 0.07893497,
0.12789202],
[ 0.28086119, 0.27569815, 0.08594638, 0.0178669 , 0.18063401,
0.15899337],
[ 0.26076848, 0.23664738, 0.08020603, 0.07001922, 0.1134371 ,
0.23892179],
[ 0.11943333, 0.29198961, 0.02605103, 0.26234032, 0.1351348 ,
0.16505091],
[ 0.09561176, 0.34396535, 0.0643941 , 0.16240774, 0.24206137,
0.09155967]])Running it through sess.run(tf.nn.top_k(tf.constant(a), k=3)) produces:
TopKV2(values=array([[ 0.34763842, 0.24879643, 0.12789202],
[ 0.28086119, 0.27569815, 0.18063401],
[ 0.26076848, 0.23892179, 0.23664738],
[ 0.29198961, 0.26234032, 0.16505091],
[ 0.34396535, 0.24206137, 0.16240774]]), indices=array([[3, 0, 5],
[0, 1, 4],
[0, 5, 1],
[1, 3, 5],
[1, 4, 3]], dtype=int32))Looking just at the first row we get [ 0.34763842, 0.24879643, 0.12789202], you can confirm these are the 3 largest probabilities in a. You'll also notice [3, 0, 5] are the corresponding indices.
In [23]:
def top_k(input_, rank):
input_ = tf.nn.softmax(input_)
return tf.nn.top_k(input_, rank)
with tf.Session() as sess:
saver.restore(sess, "lenet50")
top_5_new = sess.run(top_k(logits, 3), feed_dict={x: new_image_together})
In [24]:
k_prob = top_5_new[0]
k_label = top_5_new[1]
def show_result(image,label,k):
k_prob = k[0]
k_label = k[1]
index =len(label)
for i in range(index):
result_prob=k_prob[i]
result_label=[label_text[str(text)] for text in k_label[i]]
y_pos = np.arange(len(k_label[i]))
fig = plt.figure(figsize=(4, 1))
plt.subplot(1,2,1)
plt.title(label_text[str(label[i])])
plt.imshow(image[i])
plt.subplot(1,2,2)
plt.title("result")
plt.barh(y_pos, result_prob, align='center', alpha=0.4)
plt.yticks(y_pos, result_label)
plt.xlim([0,1])
plt.tick_params(labelright='on', labelleft="off")
plt.show()
# print(i+1,".png","prediction: ")
# print("1st: ", label_text[str(k_label[i][0])])
In [25]:
show_result(new_image_together,label_for_new_image,top_5_new)
Answer:
The result after soft_max and top_k shows which three is the correct prediection and what other result when the prediction wrong. Look like the the model prediection is really strong about what it predicts, every prediection is 100% on what it predict.
Note: Once you have completed all of the code implementations and successfully answered each question above, you may finalize your work by exporting the iPython Notebook as an HTML document. You can do this by using the menu above and navigating to \n", "File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.
In [ ]: