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.
Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the iPython Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook 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 addition to implementing code, there is a writeup to complete. The writeup should be completed in a separate file, which can be either a markdown file or a pdf document. There is a write up template that can be used to guide the writing process. Completing the code template and writeup template will cover all of the rubric points for this project.
The rubric contains "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. The stand out suggestions are optional. If you decide to pursue the "stand out suggestions", you can include the code in this Ipython notebook and also discuss the results in the writeup file.
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
# TODO: Fill this in based on where you saved the training and testing data
training_file = 'trafficsign/train.p'
validation_file='trafficsign/valid.p'
testing_file = 'trafficsign/test.p'
with open(training_file, mode='rb') as f:
train = pickle.load(f)
with open(validation_file, mode='rb') as f:
valid = pickle.load(f)
with open(testing_file, mode='rb') as f:
test = pickle.load(f)
X_train, y_train = train['features'], train['labels']
X_valid, y_valid = valid['features'], valid['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 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. Use python, numpy and/or pandas methods to calculate the data summary rather than hard coding the results. For example, the pandas shape method might be useful for calculating some of the summary results.
In [2]:
### Replace each question mark with the appropriate value.
### Use python, pandas or numpy methods rather than hard coding the results
# TODO: Number of training examples
n_train = X_train.shape[0]
# TODO: Number of validation examples
n_validation = X_valid.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.shape[1],X_train.shape[2],X_train.shape[3])
# TODO: How many unique classes/labels there are in the dataset.
n_classes = len(set(y_test))
print("Number of training examples =", n_train)
print("Number of testing examples =", n_test)
print("Number of validatio examples =", n_validation)
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. It can be interesting to look at the distribution of classes in the training, validation and test set. Is the distribution the same? Are there more examples of some classes than others?
In [3]:
### Data exploration visualization code goes here.
### Feel free to use as many code cells as needed.
import matplotlib.pyplot as plt
import pandas as pd
# Visualizations will be shown in the notebook.
%matplotlib inline
pd.DataFrame.hist(pd.DataFrame(y_train))
plt.title('Distribution of the different classes in the Training data ')
pd.DataFrame.hist(pd.DataFrame(y_valid))
plt.title('Distribution of the different classes in the Validation data ')
pd.DataFrame.hist(pd.DataFrame(y_test))
plt.title('Distribution of the different classes in the Test data ')
plt.plot()
Out[3]:
In [4]:
import numpy as np
%matplotlib inline
signNames = pd.read_csv("signnames.csv")
fig = plt.figure(figsize=(32,32), tight_layout={'h_pad':4})
for index in range(0,43,1):
image_index = (np.nonzero(y_train==index))[0][0]
ax=plt.subplot(11,4,index+1)
ax.imshow(X_train[image_index],interpolation=None)
ax.set_title( str(list(((signNames[signNames['ClassId']==index]))['SignName'])) )
plt.show()
Design and implement a deep learning model that learns to recognize traffic signs. Train and test your model on the German Traffic Sign Dataset.
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!
With the LeNet-5 solution from the lecture, you should expect a validation set accuracy of about 0.89. To meet specifications, the validation set accuracy will need to be at least 0.93. It is possible to get an even higher accuracy, but 0.93 is the minimum for a successful project submission.
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.
Minimally, the image data should be normalized so that the data has mean zero and equal variance. For image data, (pixel - 128)/ 128 is a quick way to approximately normalize the data and can be used in this project.
Other pre-processing steps are optional. You can try different techniques to see if it improves performance.
Use the code cell (or multiple code cells, if necessary) to implement the first step of your project.
In [5]:
#for gray_sclae images
import cv2
def gray(image):
return cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
def RGB_gray(X_data):
X_out_data=np.zeros((len(X_data),32,32,1))
for i in range(len(X_data)):
X_out_data[i,:,:,0]=gray(X_data[i,:,:,:])
return X_out_data
def norm(image):
dest = np.zeros((32,32,1))
return cv2.normalize(image,dest ,alpha=0, beta=1, norm_type=cv2.NORM_MINMAX, dtype=cv2.CV_32F)
def normalize(X_data):
X_out_data=np.zeros((len(X_data),32,32,X_data.shape[3]))
for i in range(len(X_data)):
X_out_data[i,:,:,0]=norm(X_data[i,:,:,0])
return X_out_data
train_gray = RGB_gray(X_train)
valid_gray = RGB_gray(X_valid)
test_gray = RGB_gray(X_test)
X_train = (train_gray-128)/128
X_valid = (valid_gray-128)/128
X_test = (test_gray-128)/128
#X_train = normalize(train_gray)
#X_valid = normalize(valid_gray)
#X_test = normalize(test_gray)
In [6]:
import numpy as np
%matplotlib inline
signNames = pd.read_csv("signnames.csv")
fig = plt.figure(figsize=(32,32), tight_layout={'h_pad':4})
for index in range(0,43,1):
image_index = (np.nonzero(y_train==index))[0][0]
ax=plt.subplot(11,4,index+1)
ax.imshow(np.squeeze(X_train[image_index]),interpolation=None)
ax.set_title( str(list(((signNames[signNames['ClassId']==index]))['SignName'])) )
plt.show()
In [7]:
import tensorflow as tf
from tensorflow.contrib.layers import flatten
EPOCHS = 20
BATCH_SIZE=128
In [8]:
def LaNet(x):
mu = 0
sigma = 0.1
#Every x is a batch with shape (128,32,32,1)
#Fallowing LANET Architecture after first convolution the ouput will be (128,28,28,6).
#The filter will be of size (5,5,1,6) so conv1_W = (5,5,1,6) and conv1_b = 6
tf.contrib.layers.xavier_initializer()
conv1_W = tf.Variable(tf.truncated_normal(shape=(5, 5,1 , 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
# The activation function is relu
conv1 = tf.nn.relu(conv1)
# We will apply max_pooling after applying max_pool our dimensions are (128,14,14,6)
conv1 = tf.nn.max_pool(conv1, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID', name='convolution1')
#2nd convolution
#after applyig 2nd convolution our output should be (128,10,10,16)
#Here we have (128,14,14,6) as input so the conv2_w =(5,5,6,16) and conv2_b=(16)
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
#activation
conv2 = tf.nn.relu(conv2)
# we apply max_pooling to the (128,10,10,16) to make it (128,5,5,16)
conv2 = tf.nn.max_pool(conv2, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='VALID', name='convolution2')
# There will be 16 feature maps each with dimesion 5*5
#print(conv2)
# 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. this will convert (128,5,5,16) to (128,400)
fc0 = flatten(conv2)
h_fc0_drop = tf.nn.dropout(fc0, keep_prob)
#print(fully_connected_layer0)
#This fully connected layer is connected to next layer with 120 neurons. So, the number of weights will be (400,120)
fc1_W = tf.Variable(tf.truncated_normal(shape=(400, 120), mean = mu, stddev = sigma))
fc1_b = tf.Variable(tf.zeros(120))
fc1 = tf.matmul(h_fc0_drop, fc1_W) + fc1_b
fc1 = tf.nn.relu(fc1)
# fully_connected_layer1 is of dimension(400,120). fully_connected_layer2 will have 84 activations
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
fc2 = tf.nn.relu(fc2)
#fully_connected_layer1 is of dimension(128,84). fully_connected_layer3 will have 43 activations
fc3_W = tf.Variable(tf.truncated_normal(shape=(84, 43), mean = mu, stddev = sigma))
fc3_b = tf.Variable(tf.zeros(43))
logits = tf.matmul(fc2, fc3_W) + fc3_b
return logits
In [9]:
x = tf.placeholder(tf.float32,(None,32,32,1))
y=tf.placeholder(tf.int32,(None))
one_hot_y = tf.one_hot(y,43)
In [10]:
rate=0.001
keep_prob = tf.placeholder(tf.float32)
logits = LaNet(x)
CrossEntropy = tf.nn.softmax_cross_entropy_with_logits(labels=one_hot_y,logits=logits)
Loss = tf.reduce_mean(CrossEntropy)
optimize = tf.train.AdamOptimizer(learning_rate=rate).minimize(Loss)
In [11]:
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, keep_prob:1.0})
total_accuracy += (accuracy * len(batch_x))
return total_accuracy / num_examples
In [12]:
from sklearn.utils import shuffle
with tf.Session() as sess:
sess.run(tf.global_variables_initializer())
num_examples = len(X_train)
print("Training...")
print()
for i in range(EPOCHS):
X_train, y_train = shuffle(X_train, y_train)
for offset in range(0, num_examples, BATCH_SIZE):
end = offset + BATCH_SIZE
batch_x, batch_y = X_train[offset:end], y_train[offset:end]
sess.run(optimize, feed_dict={x: batch_x, y: batch_y, keep_prob:0.6})
training_accuracy = evaluate(X_train, y_train)
validation_accuracy = evaluate(X_valid, y_valid)
print("EPOCH {} ...".format(i+1))
print("Training Accuracy = {:.3f}".format(training_accuracy))
print()
print("Validation Accuracy = {:.3f}".format(validation_accuracy))
print()
saver.save(sess, './lenet')
print("Model saved")
The model saved will be used on test data.
In [13]:
with tf.Session() as sess:
saver.restore(sess,tf.train.latest_checkpoint('.'))
test_accuracy = evaluate(X_test,y_test)
print(test_accuracy)
To give yourself more insight into how your model is working, download at least five pictures of German traffic signs from the web and use your model to predict the traffic sign type.
You may find signnames.csv useful as it contains mappings from the class id (integer) to the actual sign name.
In [14]:
import os
import matplotlib.image as mpimg
import cv2
test_image =[]
def Resize(img):
#resize image into (32,32) dimension
resize_img = cv2.resize(img,(32,32))
#Convert to 3 channels
#processed_img = cv2.cvtColor(resize_img, cv2.COLOR_BGRA2BGR)
test_image.append(resize_img)
return resize_img
#fig = plt.figure(figsize=(32,32), tight_layout={'h_pad':4})
i = 0
for file in os.listdir('testImages'):
if '.jpg' in file:
plt.figure(figsize=(1,1))
img = mpimg.imread('testImages/' + file)
plt.title(file)
plt.imshow(Resize(img))
plt.figure()
plt.title(file+(' Original'))
plt.imshow(img)
i+=1
test_image=normalize(RGB_gray(np.array(test_image)))