def game_details(id):
with con:
cur = con.cursor()
cur.execute("SELECT * from game_details WHERE Steam_ID=?", (id, ))
row = cur.fetchone()
x = row[9].split("$")
z = [i.split(":") for i in x]
zz = [j for j in z if len(j) == 2]
system_requirements = dict(zz)
details = {
"Name": row[1],
"Genre": row[2],
"Release Date": row[3],
"Publisher": row[4],
"Languages": row[6],
"Description": row[7],
"images": get_images(id),
"graphs": get_graphs(id)
}
print(details)
con.commit()
return render_template('game-details.html',
details=details,
sr=system_requirements,
reviews=get_reviews(id))
def render_pic():
if os.path.exists('images'):
remove_contents('images')
else:
os.mkdir('images')
text = request.form['text']
key_words = select_key_words(text)
list_of_all_img = []
for word in key_words:
list_of_img = get_images(word)
for name in list_of_img:
list_of_all_img.append(name)
print(list_of_all_img)
return render_template('test.html', list=list_of_all_img)
def project_structure(text_music_name,text_artist_name,image_type,op_lyric,op_deepDream,deepDream_format):
global current_job
shutil.rmtree('/code/flask/music', ignore_errors=True)
shutil.rmtree('/code/flask/imagens', ignore_errors=True)
json_code = get_lyric_videoLink(text_music_name,text_artist_name)
json_code['MusicPath'] = download_song(json_code['VideoID'])
print '\n\nmusica baixada\n\n'
json_code['Subtitle'] = get_images(json_code['Subtitle'],image_type)
print '\n\nimagens pegadas\n\n'
json_code['Subtitle'] = improve_subtitle(json_code['Subtitle'])
print '\n\ntimestamps modificado\n\n'
if op_deepDream:
if deepDream_format == '1':
json_code['Subtitle'] = dreamImage(json_code['Subtitle'])
if op_lyric:
video_name = make_videoDeep_lyric(json_code,text_music_name,True)
else:
video_name = make_videoDeep(json_code,text_music_name,True)
elif deepDream_format == '5':
json_code['Subtitle'] = dreamImage_5(json_code['Subtitle'])
if op_lyric:
video_name = make_videoDeep_lyric(json_code,text_music_name,False)
else:
video_name = make_videoDeep(json_code,text_music_name,False)
else:
json_code['Subtitle'] = dreamImage_10(json_code['Subtitle'])
if op_lyric:
video_name = make_videoDeep_lyric(json_code,text_music_name,False)
else:
video_name = make_videoDeep(json_code,text_music_name,False)
else:
if op_lyric:
video_name = make_video_lyric(json_code,text_music_name)
else:
video_name = make_video(json_code,text_music_name)
print '\n\nclipe feito\n\n'
return video_name
# imgs_rotated = []
# imgs_shape = []
# for image in images:
# print('Straightening {}'.format(image))
# img_rotated_cut = cut_out_corrected_img.cut_out_corrected_img(image)
# cv2.imwrite(r'./corrected_after/'+image.rpartition('/')[-1].rpartition('.')[-3][-1]+'.jpg', img_rotated_cut)
# imgs_rotated.append(img_rotated_cut)
# imgs_shape.append(img_rotated_cut.shape)
# column = 5
# img_joined = show_images.show_images(imgs_rotated, imgs_shape, column, alignment='left')
# img_ori = cv2.imread(leaf_split_before)
# # Get Corrected_after Leaves Begin
images = sorted(get_images.get_images(r'./corrected_after/'))
edges_canny = []
edges_equalized = []
edges_canny_shape = []
edges_equalized_shape = []
for image in images:
img, img_equalized, edge_canny, edge_equalized = local_enhancement.local_enhancement(image)
edges_canny.append(edge_canny)
# edges_equalized.append(edge_equalized)
edges_canny_shape.append(edge_canny.shape)
# edges_equalized_shape.append(edge_equalized.shape)
# plt.imshow(edge_canny, plt.cm.gray)
# plt.show()
# 2/9/2017 #
# #
# Data: ImageNet #
# ------------------------------------------------------------------------ #
import tensorflow as tf
import matplotlib.pyplot as plt
import numpy as np
from get_images import get_images
from DeepLearning.deep_learning import learning_rate
from Binary_Network import ResNet, train_loss
import _init_
cnn = ResNet()
data = get_images()
# tf.reset_default_graph()
x = tf.placeholder(tf.float32, [
None, _init_.input_image[0], _init_.input_image[1], _init_.input_image[2]
])
y = tf.placeholder(tf.float32, [None, _init_.classes_numbers])
# y = tf.placeholder(tf.int32, [None]) # y_5
fc_out, conv5 = cnn(x, scope='resnet')
train_step, acc_1 = train_loss(fc_out, y)
config = tf.ConfigProto(allow_soft_placement=True)
gpu_options = tf.GPUOptions(per_process_gpu_memory_fraction=0.7)
config.gpu_options.allow_growth = True
with tf.Session(config=config) as sess:
sess.run(tf.global_variables_initializer())
Acc_1 = []
import numpy as np import time import pickle from sklearn.svm import LinearSVC #from sklearn.tree import DecisionTreeClassifier #from sklearn.model_selection import GridSearchCV #from sklearn.model_selection import train_test_split from sklearn.cross_validation import train_test_split from sklearn.preprocessing import StandardScaler from sklearn.utils import shuffle from functions import extract_features from get_images import get_images # load images cars, notcars = get_images() #Feature Parameters colorspace = 'YCrCb' # Can be RGB, HSV, LUV, HLS, YUV, YCrCb orient = 9 pix_per_cell = 8 cell_per_block = 1 hog_channel = "ALL" # Can be 0, 1, 2, or "ALL" spatial = 32 histbin = 32 spatial_feat=True hist_feat=True hog_feat=True #Extra image features t=time.time() X = []
def get_image_class(path):
get_images(path)
path = get_path(path)
images_with_tags = get_prediction(model, imagenet_class_mapping, path)
print(images_with_tags)
generate_html(images_with_tags)
''' Imports '''
import get_images
import get_landmarks
from sklearn.neighbors import KNeighborsClassifier
import numpy as np
import matplotlib.pyplot as plt
from sklearn.naive_bayes import GaussianNB
''' Load the data and their labels '''
image_directory = 'H:\\Fall 2020\\biometrics\\proj\\RonKauerTestFrames'
X, y = get_images.get_images(image_directory)
''' Get distances between face landmarks in the images '''
# get_landmarks(images, labels, save_directory="", num_coords=5, to_save=False)
X, y = get_landmarks.get_landmarks(X, y, 'landmarks/', 5, False)
''' kNN classification treating every sample as a query'''
# initialize the classifie
knn_accuracy = []
NB_accuracy = []
print()
print("KNN")
for a in [1, 3, 5, 7]:
knn = KNeighborsClassifier(n_neighbors=a, metric='euclidean')
num_correct = 0
labels_correct = []
num_incorrect = 0
labels_incorrect = []
for i in range(0, len(y)):
query_img = X[i, :]
query_label = y[i]
return client
if __name__ == '__main__':
client = get_client(
project_id='sacred-reality-201417',
registry_id='robocar-ai',
device_id='donkey',
private_key_file='keys/rsa_private.pem',
algorithm='RS256',
mqtt_bridge_hostname='mqtt.googleapis.com',
mqtt_bridge_port=443,
cloud_region='us-central1',
ca_certs='keys/roots.pem')
client.loop_start()
image_file_names = get_images()
for image in image_file_names:
with open(image, 'rb') as f:
image = Image()
image.data = f.read()
image.name = 'test'
res = client.publish('/devices/donkey/events', image.SerializeToString())
print('{} is published {}'.format(image, res.is_published()))
res.wait_for_publish()
sleep(5) # Time in seconds.
# run = True
# while run:
# client.loop()
# client.
from get_images import get_images
datadirs=['/notebooks/udacity/new_training/map1_backward/',
'/notebooks/udacity/new_training/map1_forward/',
'/notebooks/udacity/new_training/map1_recovery_backward/',
'/notebooks/udacity/new_training/map1_recovery_forward/',
'/notebooks/udacity/new_training/map2_forward/',
'/notebooks/udacity/new_training/map2_backward/',
'/notebooks/udacity/new_training/map2_recovery_forward/',
'/notebooks/udacity/new_training/map2_recovery_backward/',
'/notebooks/udacity/new_training/map1_error_correction/',
'/notebooks/udacity/new_training/map2_error_correction/'
]
images=get_images(datadirs,0.08)
image_names_full, y_data_full = images.img.values, images.real.values
#preprocessing function
def proc_img(img): # input is 160x320x3
img = img[59:138:2, 0:-1:2, :] # select vertical region and take each second pixel to reduce image dimensions
img = (img / 127.5) - 1.0 # normalize colors from 0-255 to -1.0 to 1.0
return img # return 40x160x3 image
#generating train/valid sets
names_train, names_valid, y_train, y_valid = train_test_split(image_names_full, y_data_full, test_size=0.02,\
random_state=0)
#for each image adding inverse image in train set
inverse_train=[0 for i in y_train]+[1 for i in y_train]