style: reformatted the python files of the code

ksakmann · Feb 23, 2017 · 85b8814 · 85b8814
1 parent 18e1c68
commit 85b8814
Show file tree

Hide file tree

Showing 4 changed files with 0 additions and 105 deletions.
diff --git a/HOG_Classify.py b/HOG_Classify.py
@@ -1,8 +1,3 @@
-
-# coding: utf-8
-
-# In[1]:
-
 import matplotlib.image as mpimg
 import matplotlib.pyplot as plt
 import numpy as np
@@ -20,8 +15,6 @@
 from lesson_functions import *
 
 
-# In[2]:
-
 # Load the training validation and test data
 data_file = 'data.p'
 with open(data_file, mode='rb') as f:
@@ -34,8 +27,6 @@
 notcars_test  = data['notcars_test']
 
 
-# In[3]:
-
 #helper function to extract features from files
 
 def get_features(files, color_space='RGB', spatial_size=(32, 32),
@@ -55,8 +46,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
     return features
 
 
-# In[4]:
-
 color_space = 'HLS' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb
 spatial_size = (16, 16)
 hist_bins = 32
@@ -86,8 +75,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 print(round(t2-t, 2), 'Seconds to extract HOG,spatial and color features...')
 
 
-# In[5]:
-
 # Create an array stack of feature vectors
 X = np.vstack((cars_train_feat,cars_val_feat,cars_test_feat,
                notcars_train_feat,notcars_val_feat,notcars_test_feat)).astype(np.float64)                        
@@ -97,8 +84,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 scaled_X = X_scaler.transform(X)
 
 
-# In[6]:
-
 cars_ntrain=len(cars_train_feat)
 cars_nval=len(cars_val_feat)
 cars_ntest=len(cars_test_feat)
@@ -128,8 +113,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 X_test,y_test = shuffle(X_test,y_test,random_state=42)
 
 
-# In[7]:
-
 print('Using:',orient,'orientations',pix_per_cell,'pixels per cell and', cell_per_block,'cells per block')
 print('Feature vector length:', len(X_train[0]))
 # Use a linear SVC 
@@ -156,8 +139,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 print(round(t2-t, 5), 'Seconds to predict', n_predict,'labels with SVC')
 
 
-# In[18]:
-
 font_size=15
 f, axarr = plt.subplots(4, 7,figsize=(20,10))
 f.subplots_adjust(hspace=0.2, wspace=0.05)
@@ -227,8 +208,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 #plt.savefig('./images/HOG_features_YCrCb.png')
 
 
-# In[9]:
-
 # plot false positives/negatives
 font_size=15
 preds = svc.predict(cars_val_feat)
@@ -253,8 +232,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 #plt.savefig('./images/false_negatives.png')
 
 
-# In[10]:
-
 fig, axes = plt.subplots(2,10,figsize=(20,5))
 fig.subplots_adjust(hspace=0.2, wspace=0.05)
 
@@ -275,8 +252,6 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 #plt.savefig('./images/false_positives.png')
 
 
-# In[11]:
-
 # Save the data for easy access
 pickle_file = 'ProcessedData.p'
 print('Saving data to pickle file...')
@@ -325,7 +300,3 @@ def get_features(files, color_space='RGB', spatial_size=(32, 32),
 print('Data cached in pickle file.')
 
 
-# In[ ]:
-
-
-
diff --git a/exploration.py b/exploration.py
@@ -1,8 +1,3 @@
-
-# coding: utf-8
-
-# In[1]:
-
 import matplotlib.image as mpimg
 import matplotlib.pyplot as plt
 import numpy as np
@@ -16,8 +11,6 @@
 from sklearn.model_selection import train_test_split
 
 
-# In[2]:
-
 # Loading + training,validation test split
 cars0 = glob.glob('./vehicles/GTI_Far/*.png')
 cars1 = glob.glob('./vehicles/GTI_MiddleClose/*.png')
@@ -46,13 +39,6 @@
 notcars_test = notcars1[L2[5]:]
 
 
-# In[3]:
-
-L1,L2
-
-
-# In[4]:
-
 print('Number of samples in cars training set: ', len(cars_train))
 print('Number of samples in notcars training set: ', len(notcars_train))
 
@@ -63,10 +49,6 @@
 print('Number of samples in notcars test set: ',len(notcars_test))
 
 
-# **Conclusion: The two classes are more or less balanced.**
-
-# In[5]:
-
 # Save the data for easy access
 pickle_file = 'data.p'
 print('Saving data to pickle file...')
@@ -89,8 +71,6 @@
 print('Data cached in pickle file.')
 
 
-# In[6]:
-
 data_file = 'data.p'
 with open(data_file, mode='rb') as f:
     data = pickle.load(f)
@@ -101,17 +81,11 @@
 cars_test = data['cars_test']
 notcars_test = data['notcars_test']
 
-
-# In[7]:
-
 i=22
 a_car = plt.imread(cars_train[i])
 not_a_car = plt.imread(notcars_train[i])
 cars_train[i],notcars_train[i]
 
-
-# In[8]:
-
 font_size=30
 f, (ax1, ax2) = plt.subplots(1, 2, figsize=(20,10))
 ax1.imshow(a_car)
@@ -123,8 +97,6 @@
 plt.show()
 
 
-# In[9]:
-
 f, (ax1, ax2) = plt.subplots(1, 2, figsize=(20,10))
 f, (ax1, ax2) = plt.subplots(1, 2, figsize=(20,10))
 ax1.imshow(a_car)
@@ -136,7 +108,3 @@
 plt.savefig('./images/car_notcar.png')
 
 
-# In[ ]:
-
-
-
diff --git a/images/car_notcar.png b/images/car_notcar.png
diff --git a/search_classify.py b/search_classify.py
@@ -1,8 +1,3 @@
-
-# coding: utf-8
-
-# In[1]:
-
 import matplotlib.image as mpimg
 import matplotlib.pyplot as plt
 import numpy as np
@@ -25,8 +20,6 @@
 from scipy.ndimage.measurements import label
 
 
-# In[2]:
-
 # Load the classifier and parameters
 data_file = 'ClassifierData.p'
 with open(data_file, mode='rb') as f:
@@ -46,13 +39,7 @@
 hog_feat = data['hog_feat']
 
 
-# In[3]:
-
 images = sorted(glob.glob('test_images/out*.png'))
-#images
-
-
-# In[4]:
 
 #image = mpimg.imread('bbox-example-image.png')
 # Uncomment the following line if you extracted training
@@ -64,17 +51,6 @@ def search_all_scales(image):
 
     hot_windows = []
     all_windows = []
-    #Y_start_stop =[[300,460]]
-    #XY_window = [(150,150)]
-    #X_start_stop =[[None,None]]    
-    #X_start_stop =[[None,None],[None,None],[None,None]]    
-    #Y_start_stop =[[390,440],[400,560],[400,560]]    
-    #XY_window = [(80,80),(110,110),(130,130)]
-
-    X_start_stop =[[None,None],[None,None]]    
-    Y_start_stop =[[390,470],[390,500]]    
-    XY_window = [(64,64),(110,110)]
-    XY_overlap=[(0.75, 0.75),(0.75, 0.75)]
 
     X_start_stop =[[None,None],[None,None],[None,None],[None,None]]
     w0,w1,w2,w3 = 240,180,120,70
@@ -102,8 +78,6 @@ def search_all_scales(image):
     return hot_windows,all_windows
 
 
-# In[5]:
-
 for file in images[43:44]:
     image = mpimg.imread(file)
     #image = image.astype(np.float32)/255
@@ -143,8 +117,6 @@ def search_all_scales(image):
     #plt.savefig('images/sliding_windows.png')
 
 
-# In[6]:
-
 file0 = images[10]
 file1 = images[70]
 file2 = images[83]
@@ -168,12 +140,10 @@ def search_all_scales(image):
     ax.set_yticks([])     
 
 plt.show()
-#plt.savefig('images/detection_example.png')
 
 
 # ## Video Pipeline
 
-# In[7]:
 
 # Define a class to receive the characteristics of bounding box detections
 class BoundingBoxes:
@@ -228,7 +198,6 @@ def apply_threshold(heatmap, threshold):
     return heatmap
 
 
-# In[8]:
 
 images = sorted(glob.glob('./heatmap/*.png'))
 boxes = BoundingBoxes(n=6)
@@ -280,15 +249,12 @@ def apply_threshold(heatmap, threshold):
 plt.show()
 
 
-# In[9]:
-
 labels = label(heatmap)
 print(labels[1], 'cars found')
 plt.imshow(labels[0], cmap='gray')
 plt.show()
 
 
-# In[10]:
 
 def draw_labeled_bboxes(img, labels):
     # Iterate through all detected cars
@@ -314,7 +280,6 @@ def draw_labeled_bboxes(img, labels):
 plt.show()
 
 
-# In[13]:
 
 boxes = BoundingBoxes(n=30)
 
@@ -334,7 +299,6 @@ def process_image(image):
     return window_image
 
 
-# In[14]:
 
 out_dir='./output_images/'
 inpfile='project_video.mp4'
@@ -346,8 +310,6 @@ def process_image(image):
 
 # ## Images for Readme
 
-# In[ ]:
-
 heatmap_images = sorted(glob.glob('./heatmap/*.png'))
 boxes = BoundingBoxes(n=6)
 
@@ -393,17 +355,13 @@ def process_image(image):
 plt.savefig('images/heatmap.png')    
 
 
-# In[ ]:
-
 plt.figure()
 labels = label(final_heatmap)
 print(labels[1], 'cars found')
 plt.imshow(labels[0], cmap='gray')
 plt.savefig('images/labels.png')
 
 
-# In[ ]:
-
 # Read in the last image above
 image = mpimg.imread(image_seq[-1])
 # Draw bounding boxes on a copy of the image
@@ -416,7 +374,5 @@ def process_image(image):
 plt.savefig('images/bounding_boxes.png')
 
 
-# In[ ]:
-