-
Notifications
You must be signed in to change notification settings - Fork 0
/
load_data.py
152 lines (133 loc) · 4.93 KB
/
load_data.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
import csv, cv2
import numpy as np
from sklearn.utils import shuffle
# Visualize data
import matplotlib.pyplot as plt
import random
def visualize_data(X_data, y_data, title="No Title", gray_scale=False, limit=15, isRand=True):
fig, axs = plt.subplots(3, 5, figsize=(15, 6))
fig.subplots_adjust(hspace = .2, wspace=.001)
axs = axs.ravel()
for i in range(15):
axs[i].axis('off')
for i in range(limit):
index = i
if isRand == True:
index = random.randint(0, len(X_data) - 1)
image = X_data[index]
if gray_scale == True:
axs[i].imshow(image.squeeze(), cmap='gray')
else:
axs[i].imshow(image)
axs[i].set_title(y_data[index])
fig.canvas.set_window_title(title)
plt.show()
def pre_process(X_data, multiple=True):
# crop out the sky and car hood
print("> Cropping Image")
if multiple==True:
X_data = X_data[:, 75:-25]
else:
X_data = X_data[75:-25]
# normalize images
print("> Normalizing Image")
X_data = X_data / 255.0 - 0.5
# gray scale images
print("> Gray Scaling Image")
if multiple==True:
X_data = np.sum(X_data / 3, axis=3, keepdims=True)
else:
X_data = np.sum(X_data / 3, axis=2, keepdims=True)
return X_data
def augment_data(X_data, y_data):
X_output, y_output = [], []
for image, measurement in zip(X_data, y_data):
# save normal image
X_output.append(cv2.cvtColor(image, cv2.COLOR_BGR2RGB))
y_output.append(measurement)
# flip image horizontally if magnitude is > 0.33
if abs(measurement) > 0.33:
flipped_image = cv2.flip(image, 1)
X_output.append(cv2.cvtColor(flipped_image, cv2.COLOR_BGR2RGB))
y_output.append(measurement * -1)
return np.array(X_output), np.array(y_output)
def read_csv():
# Data Preparation
lines = []
with open('./data/driving_log.csv') as csv_file:
reader = csv.reader(csv_file)
next(reader, None) # skip the headers
for each_line in reader:
lines.append(each_line)
# gather data metrics
hash = {}
hash["num_samples"] = len(lines)
hash["input_shape"] = cv2.imread("./data/" + lines[0][0].strip()).shape
return lines, hash
def generator(samples, batch_size=1):
while 1: # Loop forever so the generator never terminates
shuffle(samples)
for offset in range(0, len(samples), batch_size):
batch_samples = samples[offset:offset+batch_size]
# Load Images
images = []
correction = 0.1
measurements = []
# print("> Loading Images")
for line in batch_samples:
# load left, right and center image
for index in range(3):
image_path = "./data/" + line[index].strip()
image = cv2.imread(image_path)
images.append(image)
measurement = float(line[3])
# if index == 1: # add correction to left image
# measurement = measurement + correction
# elif index == 2: # add correction to right image
# measurement = measurement - correction
# else:
measurement *= 6
measurements.append(measurement)
# convert to numpy array
images = np.array(images)
measurements = np.array(measurements)
# pre process images
# images = pre_process(images)
# Augment Data
images, measurements = augment_data(images, measurements)
# Visualize Data
# visualize_data(images, measurements, title="Images", gray_scale=True)
yield shuffle(images, measurements)
def load_data():
# Data Preparation
lines, _ = read_csv()
# Load Images
images = []
correction = 0.2
measurements = []
print("> Loading Images")
for line in lines:
# load left, right and center image
for index in range(3):
image_path = "./data/" + line[index].strip()
image = cv2.imread(image_path)
images.append(image)
measurement = float(line[3])
if index == 1: # add correction to left image
measurement = measurement + correction
elif index == 2: # add correction to right image
measurement = measurement - correction
else:
measurement *= 1
measurements.append(measurement)
# convert to numpy array
images = np.array(images)
measurements = np.array(measurements)
# pre process images
# images = pre_process(images)
# Augment Data
print("> Augmenting Data")
images, measurements = augment_data(images, measurements)
# Visualize Data
# visualize_data(images, measurements, title="Images")
return images, measurements