def load_data(self, source=1):
if source == 1:
self.numCategory, self.label_map, self.inv_label_map, self.x_train, self.y_train, self.x_val, self.y_val, self.x_test, self.y_test = \
preprocess.processing(
input_dir='../input/Category and Attribute Prediction Benchmark/Img',
labels_dir=r'../list_attr_cloth.txt',
gt_dir=r'../list_attr_img.txt'
)
else:
self.numCategory, self.label_map, self.inv_label_map, self.x_train, self.y_train, self.x_val, self.y_val, self.x_test, self.y_test = \
preprocess.processing(
input_dir='../input/Category and Attribute Prediction Benchmark/Img'
)
def main(arg=None):
#load npz file.
weights = np.load('./vgg16_weights.npz')
#Weights_Tranined store the data of vgg16_weights.npz that download from internet.
#For example: As a member of Weights_Tranined, 'conv11_w' corresponds with the
# weights of the first conv3-64. 'conv11_b' corresponds with the bias
# of the first conv3-64.
w_trained = Weights_Tranined(weights)
#Build training and validation sets
data_path = os.path.join(os.getcwd(), 'fmnist')
print(data_path)
imgs, labels = pre.processing(data_path, CLASS_NUM)
train_img, train_label, validation_img, validation_label = pre.split(
imgs, labels)
print(train_img.shape)
print(train_label.shape)
train_model(t_x=train_img,
t_y=train_label,
weights=w_trained,
dprate=DROPOUT_RATE,
imgsize=IMG_SIZE,
imgchannel=IMG_CHANNEL,
batchsize=BATCH_SIZE,
train_step=TRAINING_STEP,
learningrate=LEARNING_RATE_BASE,
learningdecay=LEARNING_RATE_DECAY,
regurate=REGULARIZATION_RATE)
def build_single_index(temp_files, document, docID): """This method controls the processing of a document and the adding of the terms to the single index.""" terms = preprocess.processing(document, 'single') append_to_index(terms, docID) temp_files = check_mem_constraint(temp_files) return temp_files
def build_positional_index(temp_files, document, docID): """This method controls the processing of a document and the adding of the terms to the phrase index.""" tokens = preprocess.processing(document, 'positional') append_to_index_position(tokens, docID) global count global memory_constraint if count > memory_constraint: temp_files = write_to_temp_position(temp_files) count = 0 return temp_files
def processImage(imgFileName):
path = os.path.join('C:/Users/prashant/Desktop/third/minorProject/images/',
imgFileName)
processed = preprocess.processing(
path) #this sends image to next module for preprocessing.
print(path)
print('\n--------------PREDICTIONS------------------\n')
x = image.img_to_array(processed)
x = np.expand_dims(x, axis=0)
with graph.as_default():
set_session(sess)
prediction = model.predict(x, batch_size=None)
print("Risk factor from image = ", prediction, type(prediction))
print('\n')
value = prediction[0][0]
print(type(value))
return value
end = time.time()
print('Preprocessing finished in {} seconds'.format(end-start))
net = dan.Net(hiddenDim = 52)
net.train(train, dev, verbose = False)
torch.save(net.state_dict(), 'C:/Users/jack1/Documents/CS5304-Final-Project-Data-Science')
'''
net = dan.Net(hiddenDim=52)
net.load_state_dict(
torch.load(
'C:/Users/jack1/Documents/CS5304-Final-Project-Data-Science/net_dict.pt'
))
print(
'\n\t REAL OR FAKE: a proof-of-concept\n\tby Irene, Aayushi and Jack\n\n')
print('Enter "exit" at any time to shut down the application.')
response = 'xxx'
while not response == 'exit':
response = input('Please enter a tweet: ')
text = [response]
proc_text = [preprocess.processing(i) for i in text]
vec = [preprocess.tweet_vec(tweet, word2vec) for tweet in proc_text]
_, y_stars = net.get_eval_data(vec, mode='test')
if y_stars[0] == 0:
print('non-Disaster')
elif y_stars[0] == 1:
print('Disaster')
print('Thank you for using REAL OR FAKE!')