def prob_image(args):
list_predictions = list()
pred = Prediction(args.proto_path,args.bin_path)
max_value = 512
curr_value = 0
with open(args.images,'r') as file_image:
list_images = list()
list_good_class = list()
for line in file_image:
splitted = line.split('\t')
list_images.append(splitted[2].strip())
curr_value = curr_value + 1
if curr_value < max_value:
continue
else:
#predict using value
predictions = pred.predict_multi(list_images)
list_images = list()
list_good_class = list()
curr_value = 0
list_predictions.append(predictions)
print predictions.shape
if len(list_images) > 0:
#predict using value
predictions = pred.predict_multi(list_images)
list_predictions.append(predictions)
all_pred = np.vstack(list_predictions)
np.save('fine_sand.npy',all_pred)
np.savetxt('probabilities_kaggle_aug_fine_m.txt', all_pred, fmt='%1.9e')
def check(event=None):
show_info("")
# try:
raw_text = displayed_file.get('1.0', tk.END)
global candidate
candidate = Prediction(raw_text, jl, jl_multiClass)
global propDict
propDict = candidate.predict_technique()
if len(propDict) > 0:
displayed_file.delete('1.0', END)
global percent_of_propaganda
percent_of_propaganda = candidate.percent_of_propaganda()
# displayed_file.insert(tk.END, )
displayed_file.insert(tk.END, raw_text)
displayed_file.tags()
statistics_button.grid(row=3, column=4, padx=10, pady=10)
print(propDict)
for key in propDict:
for value in propDict[key]:
displayed_file.highlight_pattern(value, key)
tag = key
displayed_file.tag_configure(tag)
displayed_file.tag_bind(tag,
"<Enter>",
lambda event, key=key: show_info(key))
displayed_file.tag_bind(tag,
"<Leave>",
lambda event, key=key: show_info(""))
else:
show_info("No propaganda detected")
return propDict
def join_connection():
req_data = request.get_json()
conn_id = req_data['_id']
fileName = req_data['fileName']
userId = req_data['userId']
p = Prediction()
# fo = open(app.config['UPLOAD_FOLDER'] + "/" + "ad.enc",'rb').read()
# # Decrypts the data, requires a user-input password
# CSVplaintext = decrypt('USERPASSWORD', fo).decode('utf8')
# #create a temp csv-like file to pass to pandas.read_csv()
# DATA=StringIO(CSVplaintext)
# # Makes a panda dataframe with the data
# df = pd.read_excel(DATA)
# print df.head()
value = int(p.run(app.config['UPLOAD_FOLDER'] + "/Data/" + fileName))
#conn_id = '5bf8ab7c101c372f28c65c2b'
users = mongo.db.users
connections = mongo.db.connections
parser = connections.find({'_id': ObjectId(conn_id)})
for doc in parser:
connDoc = doc
currCount = connDoc['currRemainingCount']
if currCount > 0:
playerNum = connDoc['userCount'] - (currCount - 1)
print os.system('pwd')
command = 'python ./../viff-1.0/apps/sum.py --no-ssl player-' + str(
playerNum) + '.ini ' + str(value)
print command
print connections.update({'_id': ObjectId(conn_id)},
{'$inc': {
'currRemainingCount': -1
}})
#os.system(command)
proc = Popen(command,
shell=True,
stdin=None,
stdout=None,
stderr=None,
close_fds=True)
print "Status = " + str(proc)
status = 0
if status == 0:
#print output
value *= 100
users.insert({
'userId': str(userId),
'yourVal': value,
'AvgValue': 0
})
retVal = {}
retVal['sum'] = 'Will be updated soon!'
return jsonify(retVal), 200
else:
'Joining Failed', 404
else:
return 'Connection Full', 404
class ChatBotInit:
def __init__(self):
#TODO move to outside
self.model_name = CHATBOT_MODEL
self.embedding_file = 'embedding_wiki_100d_pt.txt'
self.ds = Dataset()
self.messages = self.ds.import_dataset()
self.questions = self.ds.get_questions(
self.messages) #['why?', 'how?', 'when?, ...']
self.answers = self.ds.get_answers(
self.messages) #['yes', 'no', 'maybe', ...]
self.model = load_model(CHATDATA_DIR + self.model_name)
self.pc_questions = self.ds.get_page_compute(qea=1)
self.pc_answers = self.ds.get_page_compute(qea=0)
self.tokenizer = self.ds.load_tokenizer()
self.p = Prediction(model=self.model,
messages=self.messages,
questions=self.questions,
answers=self.answers,
pc_questions=self.pc_questions,
pc_answers=self.pc_answers,
tokenizer=self.tokenizer)
def init_chat_cmd(self, run_once=False):
m = None
exit = 'exit'
print('I am ready to talk.')
print(' Tip: You can type \"exit\" (lowercase) anytime to leave.')
while (m != exit):
#TODO save conversation to file_timestamp
try:
m = input(YOU_PREFIX)
save_content_to_log_file(YOU_PREFIX + str(m))
except KeyboardInterrupt:
print(BOT_PREFIX + emergency_message())
save_content_to_log(sys.exc_info())
if m != exit:
r = str(self.p.predict(m))
print(BOT_PREFIX + r)
save_content_to_log_file(BOT_PREFIX + str(r))
if run_once == True:
m = exit
print(BOT_PREFIX + 'Bye!')
def get_conversation(self, message):
return str(self.p.predict(message))
def test_accuracy_multi(args):
# Compute Accuracy of images from input file
# Code predict many images at one time. The value is set by max_value
# Same value should be set in proto file
# It is done because of speed reason
pred = Prediction(args.proto_path,args.bin_path)
max_value = 512
curr_value = 0
list_all_result = list()
list_good_class_all = list()
with open(args.images,'r') as file_image:
list_images = list()
list_good_class = list()
for line in file_image:
splitted = line.split(' ')
#if not os.path.isfile(splitted[0].strip()):
#continue
list_good_class.append(int(splitted[1]))
list_images.append(splitted[0].strip())
curr_value = curr_value + 1
if curr_value < max_value:
continue
else:
#predict using value
predictions = pred.predict_multi(list_images)
list_all_result.append(predictions)
list_good_class_all.extend(list_good_class)
list_good_class = list()
list_images = list()
curr_value = 0
#predict last package of data, which is smaller than max_value
if len(list_images) > 0:
predictions = pred.predict_multi(list_images)
list_all_result.append(predictions)
#test loss
list_good_class_all.extend(list_good_class)
y_truth = np.asarray(list_good_class_all)
y_pred = np.vstack(list_all_result)
y_pred_label = np.argmax(y_pred, axis=1)
print y_truth.shape, y_pred.shape
print "Accuracy: ", accuracy_score(y_truth,y_pred_label)
print "Loss: ",multiclass_log_loss(y_truth,y_pred)
plot_confusion_matrix(y_truth,y_pred_label)
np.save('casia_deep.npy',y_pred)
def predict_image(model_path: str, image: np.ndarray, show: bool = False):
pred = Prediction(model_path=model_path,
output_path=None,
patch_size=train_params['patch_size'],
batch_size=train_params['batch_size'],
offset=2 * train_params["n_layers"],
n_classes=len(classes_mask.keys()),
n_filters=train_params["n_filters"],
n_layers=train_params["n_layers"])
# image = np.array(Image.open(image)).astype(np.float32) / 255
predicted = pred.__predict_image__(image)
predicted = np.argmax(predicted, axis=2)
if show:
Image.fromarray(
colorize_mask(np.dstack((predicted, predicted, predicted)),
n_classes=pred.n_classes)).show()
return predicted
def __init__(self):
#TODO move to outside
self.model_name = CHATBOT_MODEL
self.embedding_file = 'embedding_wiki_100d_pt.txt'
self.ds = Dataset()
self.messages = self.ds.import_dataset()
self.questions = self.ds.get_questions(
self.messages) #['why?', 'how?', 'when?, ...']
self.answers = self.ds.get_answers(
self.messages) #['yes', 'no', 'maybe', ...]
self.model = load_model(CHATDATA_DIR + self.model_name)
self.pc_questions = self.ds.get_page_compute(qea=1)
self.pc_answers = self.ds.get_page_compute(qea=0)
self.tokenizer = self.ds.load_tokenizer()
self.p = Prediction(model=self.model,
messages=self.messages,
questions=self.questions,
answers=self.answers,
pc_questions=self.pc_questions,
pc_answers=self.pc_answers,
tokenizer=self.tokenizer)
def __init__( self ):
"""Constructor."""
super( QMainWindow, self ).__init__( )
self.setGeometry(10, 50, 300, 150)
self.setWindowTitle("Informative or NonInformative Tweets")
self.filename = ""
self.prediction = Prediction( "../models/text.h5", "../models/old_image/damage.h5" )
self.widget = QWidget ( )
self.widget.setObjectName("widget")
self.widget.setStyleSheet("QWidget#widget{border-image:url('../images/background.jpg');}\n")
self.layout = QGridLayout( self.widget )
self.image_label = QLabel( )
self.result_label = QLabel( )
self.image_label .setScaledContents( True )
self.result_label.setScaledContents( True )
self.browse = QPushButton( "Browse" )
self.text_field = QLineEdit ( )
self.submit = QPushButton( "Submit" )
self.layout.addWidget( self.image_label )
self.layout.addWidget( self.browse )
self.layout.addWidget( self.text_field )
self.layout.addWidget( self.submit )
self.layout.addWidget( self.result_label )
self.browse.clicked.connect( self.pick_file )
self.submit.clicked.connect( self.predict )
self.setCentralWidget( self.widget )
def dispatch(values=None):
#Validate parm
if (values == None):
return {'error': 'parameter is missing'}
if (not (isinstance(values, dict))):
return {'error': 'parameter is not a dictionary'}
if (not ('op' in values)):
values['error'] = 'no op is specified'
return values
#Perform designated function
if (values['op'] == 'adjust'):
return AdjustedAltitude.dispatch(values)
elif (values['op'] == 'predict'):
return Prediction.dispatch(values)
elif (values['op'] == 'correct'):
return Correction.dispatch(values)
elif (values['op'] == 'locate'):
return values #This calculation is stubbed out
else:
values['error'] = 'op is not a legal operation'
return values
import MySQLdb.cursors
import re
import os
import cv2
import image_processing.filters as filters
import image_processing.preprocessing as preprocessing
import image_processing.to_domain as to_domain
import time
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
from predict import Prediction
predictor = Prediction("model_weights.h5")
app = Flask(__name__)
app.secret_key = 'thisissecret'
app.config['MYSQL_HOST'] = 'localhost'
app.config['MYSQL_USER'] = '******'
app.config['MYSQL_PASSWORD'] = ''
app.config['MYSQL_DB'] = 'retinadb'
mysql = MySQL(app)
predicted = {}
processed_images = {}
@app.route("/index")
mode = sys.argv[1]
if mode == "train" or mode == "onlyexport":
data_file = sys.argv[2]
parameter_file = sys.argv[3]
examples_file = sys.argv[4]
with codecs.open(parameter_file, 'r', 'utf-8') as f:
params = json.load(f)
with codecs.open(examples_file, 'r', 'utf-8') as f:
examples = json.load(f)
t = Training(data_file=data_file, examples=examples)
for p in params:
print("Trying:{}".format(p))
p["only_export"] = (mode == "onlyexport")
t.train(**p)
if mode == "predict":
vocab_file = sys.argv[2]
model_file = sys.argv[3]
text = sys.argv[4]
p = Prediction(model_file, vocab_file)
print(p.predict(text))
if mode == "predict_server":
data_file = sys.argv[2]
model_file = sys.argv[3]
p = Prediction(model_file, data_file)
db = get_database_from_file(data_file)
PredictServer(p, db)
class MainWindow( QMainWindow ):
"""Class implementing GUI"""
def __init__( self ):
"""Constructor."""
super( QMainWindow, self ).__init__( )
self.setGeometry(10, 50, 300, 150)
self.setWindowTitle("Informative or NonInformative Tweets")
self.filename = ""
self.prediction = Prediction( "../models/text.h5", "../models/old_image/damage.h5" )
self.widget = QWidget ( )
self.widget.setObjectName("widget")
self.widget.setStyleSheet("QWidget#widget{border-image:url('../images/background.jpg');}\n")
self.layout = QGridLayout( self.widget )
self.image_label = QLabel( )
self.result_label = QLabel( )
self.image_label .setScaledContents( True )
self.result_label.setScaledContents( True )
self.browse = QPushButton( "Browse" )
self.text_field = QLineEdit ( )
self.submit = QPushButton( "Submit" )
self.layout.addWidget( self.image_label )
self.layout.addWidget( self.browse )
self.layout.addWidget( self.text_field )
self.layout.addWidget( self.submit )
self.layout.addWidget( self.result_label )
self.browse.clicked.connect( self.pick_file )
self.submit.clicked.connect( self.predict )
self.setCentralWidget( self.widget )
def pick_file( self ):
"""For selecting a file."""
self.filename = QFileDialog.getOpenFileName( )[ 0 ]
if self.filename != "":
pixmap = QPixmap ( self.filename )
self.image_label.setPixmap( pixmap )
def predict( self ):
if self.filename != "":
image = cv2.imread( self.filename )
text = self.text_field.text( )
im_class, txt_class = self.prediction.run( image, text )
if im_class == 0 and txt_class == 0:
self.result_label.setText( "<font color='red'>Informative</font>" )
else:
self.result_label.setText( "<font color='green'>Non-Informative</font>" )
from numpy import array
import numpy as np
import tensorflow as tf
from tensorflow.keras.preprocessing.text import tokenizer_from_json
from predict import Prediction
import json
# Load saved tokenizer
with open('tokenizer_30000.json') as f:
data = json.load(f)
tokenizer = tokenizer_from_json(data)
max_length = 179
pred = Prediction(tokenizer, max_length)
pred.load_model()
#we were laughing , haha thats funny but then we were saying damn it all , what kind of example is this guy setting for the kids ?
#this is just great , seth shouted .
#he might be into it .
#just try to be less annoying than usual
#this may say something important
# Predict for given sequence
print(pred.predict_sequnce("just try to be less annoying than usual", 100))
