maxs = max(sensors_data)
correlations = correlation(sensors_data)
# fft_vals = fft(sensors_data)
# entropies = entropy(fft_vals)
# energies = energy(fft_vals)
return np.hstack([means, vars, mins, maxs, correlations])
def extract_features(segments):
return np.array(
[extract_features_over_segment(segment) for segment in segments])
if __name__ == '__main__':
import os
import data_collection
import preprocess
np.set_printoptions(suppress=True)
EXP_LOCATION = os.path.join('data', 'varunchicken1')
collector = data_collection.DataCollection(EXP_LOCATION)
collector.load()
segments = collector.segment()
segments = preprocess.preprocess_segments(segments[0:3])
print(extract_features(segments).shape)
index1_1 = 0 index0_2 = 0 index1_2 = 0 index0_3 = 0 index1_3 = 0 distance0 = rc.RatioCalculation() distance1 = rc.RatioCalculation() distance2 = rc.RatioCalculation() # Creating data collection / hit detection log objects sheet_index = 0 # Sheet index in Google Sheets hit_obj0 = dc.DataCollection() hit0 = hit_obj0.run_once_hit(hit_obj0.hit_count) time0 = hit_obj0.run_once_time(hit_obj0.time_count) hit_obj1 = dc.DataCollection() hit1 = hit_obj1.run_once_hit(hit_obj1.hit_count) time1 = hit_obj1.run_once_time(hit_obj1.time_count) hit_obj2 = dc.DataCollection() hit2 = hit_obj2.run_once_hit(hit_obj2.hit_count) time2 = hit_obj2.run_once_time(hit_obj2.time_count) start_time = datetime.datetime.now().replace(microsecond=0) print(start_time) with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
app.config["SESSION_PERMANENT"] = False
app.config["SESSION_TYPE"] = "filesystem"
Session(app)
# Configure CS50 Library to use SQLite database
db = SQL("sqlite:///user_info.db")
# camera configuration
camera = cv2.VideoCapture(0)
exercise = None
video_obj = VideoCamera()
# object for data collection
data_collection_obj = data_collection.DataCollection()
# success_exercise = False
@app.after_request
def add_header(r):
"""
Add headers to both force latest IE rendering engine or Chrome Frame,
and also to cache the rendered page for 10 minutes.
"""
r.headers["Cache-Control"] = "no-cache, no-store, must-revalidate"
r.headers["Pragma"] = "no-cache"
r.headers["Expires"] = "0"
r.headers['Cache-Control'] = 'public, max-age=0'
return r
import sys
sys.path.insert(0,'class_')
import nlu,tts
import random
import data_collection
# initialize the tts and the datacollection class
TTS = tts.Tts()
TTS.set_property_voice()
datacollection = data_collection.DataCollection()
class DM:
""" This is a class about dialogue management. It aims to manage all slots and intent compoted from nlu
and then return the answer to user.
"""
def __init__(self,what_film,number_of_tikets,what_time,when,location,gen_info,choice_gen):
self.intent_class = nlu.IntentCalssifier()
self.what_film = what_film
self.number_of_tikets = number_of_tikets
self.what_time = what_time
self.when = when
self.location = location
self.gen_info = gen_info
self.choice_gen = choice_gen
self.slot_booking_film = [self.what_film,self.number_of_tikets,self.what_time,self.location,self.when]
self.intents = self.intent_class.intents_class
# manage al intent for first time. It is called when the user doesn't repeat. Thus the
# robot has undestand the answer of user
def manage_first_action(self, intent):
def __init__(self, focal, pp, model_path):
self.focal = focal
self.pp = pp
self.data_collection = data_collection.DataCollection()
self.loaded_model = load_model(model_path)
def collect_tweets():
with open('config.json', 'r') as f:
config = json.load(f)
d = dc.DataCollection(config)
d.collect_tweets('output', 'a')