\u0410\u0432\u0442\u043e\u0440 – \u0421\u0443\u0445\u0430\u0447\u043e\u0432 \u0414\u0435\u043d\u0438\u0441 \u041f\u0430\u0432\u043b\u043e\u0432\u0438\u0447<\/p>\n\n\n\n
This is a work, and the author’s right to a work under international law comes into force from the moment the work is created.<\/p>\n\n\n\n
<\/p>\n\n\n\n
\u041d\u0438\u0436\u0447\u0435 \u043d\u0430\u0432\u0435\u0434\u0435\u043d\u043e \u043e\u043d\u043e\u0432\u043b\u0435\u043d\u0438\u0439 \u0432\u0430\u0440\u0456\u0430\u043d\u0442 \u043a\u043e\u0434\u0443 \u0437 \u043d\u0430\u0441\u0442\u0443\u043f\u043d\u0438\u043c\u0438 \u0434\u043e\u043f\u043e\u0432\u043d\u0435\u043d\u043d\u044f\u043c\u0438:<\/p>\n\n\n\n
python<\/p>\n\n\n\n
\u041a\u043e\u043f\u0456\u044e\u0432\u0430\u0442\u0438<\/p>\n\n\n\n
import numpy as np<\/p>\n\n\n\n
import pickle<\/p>\n\n\n\n
import os<\/p>\n\n\n\n
import time<\/p>\n\n\n\n
import cv2<\/p>\n\n\n\n
import logging<\/p>\n\n\n\n
import speech_recognition as sr<\/p>\n\n\n\n
from concurrent.futures import ThreadPoolExecutor, as_completed<\/p>\n\n\n\n
# \u041d\u0430\u043b\u0430\u0448\u0442\u0443\u0432\u0430\u043d\u043d\u044f \u043b\u043e\u0433\u0443\u0432\u0430\u043d\u043d\u044f<\/p>\n\n\n\n
logging.basicConfig(level=logging.INFO, format=’%(asctime)s [%(levelname)s] %(message)s’)<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 Neuron =======================<\/p>\n\n\n\n
class Neuron:<\/p>\n\n\n\n
def __init__(self, vector=None, memory_file=’neuron_memory.pkl’,<\/p>\n\n\n\n
consciousness_frequency=1.0, thought_speed=1.0, info_quality=1.0,<\/p>\n\n\n\n
discreteness=1.0, streamness=1.0, density=1.0, interconnectedness=1.0):<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
\u041d\u0435\u0439\u0440\u043e\u043d \u0437 \u0432\u0435\u043a\u0442\u043e\u0440\u043e\u043c \u0432\u0430\u0433, \u0432\u043d\u0443\u0442\u0440\u0456\u0448\u043d\u044c\u043e\u044e \u043f\u0430\u043c’\u044f\u0442\u0442\u044e \u0442\u0430 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u0430\u043c\u0438 \u00ab\u0441\u0432\u0456\u0434\u043e\u043c\u043e\u0441\u0442\u0456\u00bb.<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
if vector is None:<\/p>\n\n\n\n
self.vector = np.random.rand(3)<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
self.vector = np.array(vector, dtype=float)<\/p>\n\n\n\n
norm = np.linalg.norm(self.vector)<\/p>\n\n\n\n
if norm > 0:<\/p>\n\n\n\n
self.vector = self.vector \/ norm<\/p>\n\n\n\n
self.memory = {}<\/p>\n\n\n\n
self.memory_file = memory_file<\/p>\n\n\n\n
self.load_memory()<\/p>\n\n\n\n
self.connections = {}<\/p>\n\n\n\n
self.consciousness_frequency = consciousness_frequency<\/p>\n\n\n\n
self.thought_speed = thought_speed<\/p>\n\n\n\n
self.info_quality = info_quality<\/p>\n\n\n\n
self.discreteness = discreteness<\/p>\n\n\n\n
self.streamness = streamness<\/p>\n\n\n\n
self.density = density<\/p>\n\n\n\n
self.interconnectedness = interconnectedness<\/p>\n\n\n\n
self.logic_rules = {<\/p>\n\n\n\n
“fact”: [],<\/p>\n\n\n\n
“unverified”: [],<\/p>\n\n\n\n
“probable”: [],<\/p>\n\n\n\n
“a_priori”: []<\/p>\n\n\n\n
}<\/p>\n\n\n\n
logging.info(“Neuron \u0456\u043d\u0456\u0446\u0456\u0430\u043b\u0456\u0437\u043e\u0432\u0430\u043d\u043e \u0437 \u0432\u0435\u043a\u0442\u043e\u0440\u043e\u043c %s”, self.vector)<\/p>\n\n\n\n
def add_connection(self, neuron, weight_coefficient=1.0):<\/p>\n\n\n\n
self.connections[neuron] = weight_coefficient<\/p>\n\n\n\n
logging.info(“\u0414\u043e\u0434\u0430\u043d\u043e \u0437\u0432’\u044f\u0437\u043e\u043a \u0437 \u043d\u0435\u0439\u0440\u043e\u043d\u043e\u043c \u0456\u0437 \u043a\u043e\u0435\u0444\u0456\u0446\u0456\u0454\u043d\u0442\u043e\u043c %s”, weight_coefficient)<\/p>\n\n\n\n
def update_weight(self, input_vector, learning_rate=0.1):<\/p>\n\n\n\n
input_vector = np.array(input_vector, dtype=float)<\/p>\n\n\n\n
self.vector += learning_rate * input_vector<\/p>\n\n\n\n
norm = np.linalg.norm(self.vector)<\/p>\n\n\n\n
if norm > 0:<\/p>\n\n\n\n
self.vector = self.vector \/ norm<\/p>\n\n\n\n
logging.info(“\u0412\u0435\u043a\u0442\u043e\u0440 \u0432\u0430\u0433 \u043e\u043d\u043e\u0432\u043b\u0435\u043d\u043e \u0434\u043e %s”, self.vector)<\/p>\n\n\n\n
def store_memory(self, key, value, weight=1.0, category=”fact”):<\/p>\n\n\n\n
self.memory[key] = {“value”: value, “weight”: weight, “category”: category}<\/p>\n\n\n\n
if category in self.logic_rules:<\/p>\n\n\n\n
self.logic_rules[category].append(key)<\/p>\n\n\n\n
self.save_memory()<\/p>\n\n\n\n
logging.info(“\u0417\u0430\u043f\u0438\u0441\u0430\u043d\u043e \u043f\u0430\u043c’\u044f\u0442\u044c: %s -> %s”, key, value)<\/p>\n\n\n\n
def save_memory(self):<\/p>\n\n\n\n
with open(self.memory_file, ‘wb’) as f:<\/p>\n\n\n\n
pickle.dump(self.memory, f)<\/p>\n\n\n\n
logging.info(“\u041f\u0430\u043c’\u044f\u0442\u044c \u0437\u0431\u0435\u0440\u0435\u0436\u0435\u043d\u043e \u0443 \u0444\u0430\u0439\u043b %s”, self.memory_file)<\/p>\n\n\n\n
def load_memory(self):<\/p>\n\n\n\n
if os.path.exists(self.memory_file):<\/p>\n\n\n\n
with open(self.memory_file, ‘rb’) as f:<\/p>\n\n\n\n
self.memory = pickle.load(f)<\/p>\n\n\n\n
logging.info(“\u041f\u0430\u043c’\u044f\u0442\u044c \u0437\u0430\u0432\u0430\u043d\u0442\u0430\u0436\u0435\u043d\u043e \u0437 \u0444\u0430\u0439\u043b\u0443 %s”, self.memory_file)<\/p>\n\n\n\n
def activate(self, input_vector):<\/p>\n\n\n\n
input_vector = np.array(input_vector, dtype=float)<\/p>\n\n\n\n
activation_value = np.dot(self.vector, input_vector)<\/p>\n\n\n\n
for neuron, coeff in self.connections.items():<\/p>\n\n\n\n
activation_value += coeff * np.dot(neuron.vector, input_vector)<\/p>\n\n\n\n
modulation = self.consciousness_frequency * self.thought_speed * self.info_quality<\/p>\n\n\n\n
activation_value *= modulation<\/p>\n\n\n\n
logging.info(“\u0410\u043a\u0442\u0438\u0432\u0430\u0446\u0456\u044f \u043d\u0435\u0439\u0440\u043e\u043d\u0430: %s”, activation_value)<\/p>\n\n\n\n
return activation_value<\/p>\n\n\n\n
def update_consciousness_parameters(self, consciousness_frequency=None, thought_speed=None,<\/p>\n\n\n\n
info_quality=None, discreteness=None,<\/p>\n\n\n\n
streamness=None, density=None, interconnectedness=None):<\/p>\n\n\n\n
if consciousness_frequency is not None:<\/p>\n\n\n\n
self.consciousness_frequency = consciousness_frequency<\/p>\n\n\n\n
if thought_speed is not None:<\/p>\n\n\n\n
self.thought_speed = thought_speed<\/p>\n\n\n\n
if info_quality is not None:<\/p>\n\n\n\n
self.info_quality = info_quality<\/p>\n\n\n\n
if discreteness is not None:<\/p>\n\n\n\n
self.discreteness = discreteness<\/p>\n\n\n\n
if streamness is not None:<\/p>\n\n\n\n
self.streamness = streamness<\/p>\n\n\n\n
if density is not None:<\/p>\n\n\n\n
self.density = density<\/p>\n\n\n\n
if interconnectedness is not None:<\/p>\n\n\n\n
self.interconnectedness = interconnectedness<\/p>\n\n\n\n
logging.info(“\u041e\u043d\u043e\u0432\u043b\u0435\u043d\u043e \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u0438 \u0441\u0432\u0456\u0434\u043e\u043c\u043e\u0441\u0442\u0456: %s”, self.assess_state())<\/p>\n\n\n\n
def assess_state(self):<\/p>\n\n\n\n
return {<\/p>\n\n\n\n
“consciousness_frequency”: self.consciousness_frequency,<\/p>\n\n\n\n
“thought_speed”: self.thought_speed,<\/p>\n\n\n\n
“info_quality”: self.info_quality,<\/p>\n\n\n\n
“discreteness”: self.discreteness,<\/p>\n\n\n\n
“streamness”: self.streamness,<\/p>\n\n\n\n
“density”: self.density,<\/p>\n\n\n\n
“interconnectedness”: self.interconnectedness<\/p>\n\n\n\n
}<\/p>\n\n\n\n
def logical_evaluation(self, input_data):<\/p>\n\n\n\n
if isinstance(input_data, str) and “exp” in input_data.lower():<\/p>\n\n\n\n
category = “fact”<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
category = “probable”<\/p>\n\n\n\n
key = f”evaluation_{time.time()}”<\/p>\n\n\n\n
self.store_memory(key, input_data, weight=1.0, category=category)<\/p>\n\n\n\n
logging.info(“\u041b\u043e\u0433\u0456\u0447\u043d\u0430 \u043e\u0446\u0456\u043d\u043a\u0430: %s -> %s”, input_data, category)<\/p>\n\n\n\n
return category<\/p>\n\n\n\n
def associate_memories(self, key1, key2, association_strength=0.1):<\/p>\n\n\n\n
if key1 in self.memory and key2 in self.memory:<\/p>\n\n\n\n
combined_key = f”{key1}-{key2}”<\/p>\n\n\n\n
combined_value = (self.memory[key1][“value”], self.memory[key2][“value”])<\/p>\n\n\n\n
combined_weight = (self.memory[key1][“weight”] + self.memory[key2][“weight”]) * association_strength<\/p>\n\n\n\n
self.memory[combined_key] = {“value”: combined_value, “weight”: combined_weight, “category”: “associated”}<\/p>\n\n\n\n
self.save_memory()<\/p>\n\n\n\n
logging.info(“\u0410\u0441\u043e\u0446\u0456\u0430\u0446\u0456\u044f \u0441\u0442\u0432\u043e\u0440\u0435\u043d\u0430 \u043c\u0456\u0436 %s \u0442\u0430 %s”, key1, key2)<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 Perceptron =======================<\/p>\n\n\n\n
class Perceptron:<\/p>\n\n\n\n
def __init__(self, neurons=None):<\/p>\n\n\n\n
if neurons is None:<\/p>\n\n\n\n
self.neurons = []<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
self.neurons = neurons<\/p>\n\n\n\n
self.mental_constructions = {}<\/p>\n\n\n\n
logging.info(“Perceptron \u0456\u043d\u0456\u0446\u0456\u0430\u043b\u0456\u0437\u043e\u0432\u0430\u043d\u043e \u0456\u0437 %s \u043d\u0435\u0439\u0440\u043e\u043d\u0430\u043c\u0438”, len(self.neurons))<\/p>\n\n\n\n
def add_neuron(self, neuron):<\/p>\n\n\n\n
self.neurons.append(neuron)<\/p>\n\n\n\n
logging.info(“\u041d\u0435\u0439\u0440\u043e\u043d \u0434\u043e\u0434\u0430\u043d\u043e \u0434\u043e \u043f\u0435\u0440\u0446\u0435\u043f\u0442\u0440\u043e\u043d\u0430.”)<\/p>\n\n\n\n
def evaluate_network(self, input_vector):<\/p>\n\n\n\n
outputs = [neuron.activate(input_vector) for neuron in self.neurons]<\/p>\n\n\n\n
global_activation = sum(outputs) \/ len(outputs) if outputs else 0<\/p>\n\n\n\n
logging.info(“\u0413\u043b\u043e\u0431\u0430\u043b\u044c\u043d\u0430 \u0430\u043a\u0442\u0438\u0432\u0430\u0446\u0456\u044f \u043c\u0435\u0440\u0435\u0436\u0456: %s”, global_activation)<\/p>\n\n\n\n
return global_activation<\/p>\n\n\n\n
def build_mental_construction(self, construct_name, neuron_keys):<\/p>\n\n\n\n
combined_data = {}<\/p>\n\n\n\n
for neuron in self.neurons:<\/p>\n\n\n\n
for key in neuron.memory:<\/p>\n\n\n\n
if key in neuron_keys:<\/p>\n\n\n\n
combined_data[key] = neuron.memory[key]<\/p>\n\n\n\n
self.mental_constructions[construct_name] = combined_data<\/p>\n\n\n\n
logging.info(“\u041c\u0435\u043d\u0442\u0430\u043b\u044c\u043d\u0430 \u043a\u043e\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0456\u044f ‘%s’ \u043f\u043e\u0431\u0443\u0434\u043e\u0432\u0430\u043d\u0430.”, construct_name)<\/p>\n\n\n\n
return self.mental_constructions[construct_name]<\/p>\n\n\n\n
def assess_network_state(self):<\/p>\n\n\n\n
states = [neuron.assess_state() for neuron in self.neurons]<\/p>\n\n\n\n
logging.info(“\u0421\u0442\u0430\u043d \u043c\u0435\u0440\u0435\u0436\u0456: %s”, states)<\/p>\n\n\n\n
return states<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 ImageProcessor =======================<\/p>\n\n\n\n
class ImageProcessor:<\/p>\n\n\n\n
def __init__(self, process_color=True):<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
\u042f\u043a\u0449\u043e process_color=True, \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u043d\u0435 \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u044f\u0442\u044c\u0441\u044f \u0443 \u0432\u0456\u0434\u0442\u0456\u043d\u043a\u0438 \u0441\u0456\u0440\u043e\u0433\u043e.<\/p>\n\n\n\n
\u0414\u043b\u044f \u043f\u043e\u0432\u043d\u043e\u043a\u043e\u043b\u0456\u0440\u043d\u043e\u0457 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u043c\u0438 \u0431\u0443\u0434\u0435\u043c\u043e \u043e\u0431\u0447\u0438\u0441\u043b\u044e\u0432\u0430\u0442\u0438 \u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u043a\u043e\u043b\u0456\u0440 \u0437\u0430 \u043a\u0430\u043d\u0430\u043b\u0430\u043c\u0438.<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
self.process_color = process_color<\/p>\n\n\n\n
def preprocess_image(self, image_path):<\/p>\n\n\n\n
image = cv2.imread(image_path)<\/p>\n\n\n\n
if image is None:<\/p>\n\n\n\n
error_msg = “\u0417\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u043d\u0435 \u0437\u043d\u0430\u0439\u0434\u0435\u043d\u043e \u0437\u0430 \u0448\u043b\u044f\u0445\u043e\u043c: ” + image_path<\/p>\n\n\n\n
logging.error(error_msg)<\/p>\n\n\n\n
raise ValueError(error_msg)<\/p>\n\n\n\n
if self.process_color:<\/p>\n\n\n\n
# \u041d\u0435 \u043f\u0435\u0440\u0435\u0432\u043e\u0434\u0438\u043c\u043e \u0432 \u0433\u0440\u0430\u0434\u0430\u0446\u0456\u0457 \u0441\u0456\u0440\u043e\u0433\u043e \u2013 \u043f\u0440\u0430\u0446\u044e\u0454\u043c\u043e \u0437 \u043a\u043e\u043b\u044c\u043e\u0440\u043e\u0432\u0438\u043c \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f\u043c<\/p>\n\n\n\n
return image<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n\n\n\n
return gray<\/p>\n\n\n\n
def extract_color_features(self, image):<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
\u0414\u043b\u044f \u043f\u043e\u0432\u043d\u043e\u043a\u043e\u043b\u0456\u0440\u043d\u043e\u0457 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u043e\u0431\u0447\u0438\u0441\u043b\u044e\u0454\u043c\u043e \u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u043a\u043e\u043b\u0456\u0440 (B, G, R) \u043f\u043e \u0432\u0441\u0456\u0445 \u043f\u0456\u043a\u0441\u0435\u043b\u044f\u0445.<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
mean_color = np.mean(image, axis=(0, 1))<\/p>\n\n\n\n
logging.info(“\u041e\u0431\u0447\u0438\u0441\u043b\u0435\u043d\u043e \u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u043a\u043e\u043b\u0456\u0440: %s”, mean_color)<\/p>\n\n\n\n
return mean_color<\/p>\n\n\n\n
def extract_hu_moments(self, image):<\/p>\n\n\n\n
# \u042f\u043a\u0449\u043e \u043f\u0440\u0430\u0446\u044e\u0454\u043c\u043e \u0432 \u0440\u0435\u0436\u0438\u043c\u0456 \u0441\u0456\u0440\u043e\u0433\u043e, \u043c\u043e\u0436\u0435\u043c\u043e \u043e\u0431\u0447\u0438\u0441\u043b\u044e\u0432\u0430\u0442\u0438 Hu Moments<\/p>\n\n\n\n
moments = cv2.moments(image)<\/p>\n\n\n\n
huMoments = cv2.HuMoments(moments).flatten()<\/p>\n\n\n\n
huMoments = -np.sign(huMoments) * np.log10(np.abs(huMoments) + 1e-10)<\/p>\n\n\n\n
return huMoments<\/p>\n\n\n\n
def process_image(self, image_path):<\/p>\n\n\n\n
image = self.preprocess_image(image_path)<\/p>\n\n\n\n
if self.process_color:<\/p>\n\n\n\n
# \u0412\u0438\u0434\u0456\u043b\u044f\u0454\u043c\u043e \u043f\u0440\u043e\u0441\u0442\u0438\u0439 \u043a\u043e\u043b\u0456\u0440\u043d\u0438\u0439 \u0434\u0435\u0441\u043a\u0440\u0438\u043f\u0442\u043e\u0440 \u2013 \u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u043a\u043e\u043b\u0456\u0440 (B, G, R)<\/p>\n\n\n\n
features = self.extract_color_features(image)<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
preprocessed = image # \u0432\u0436\u0435 \u0432 \u0433\u0440\u0430\u0434\u0430\u0446\u0456\u044f\u0445 \u0441\u0456\u0440\u043e\u0433\u043e<\/p>\n\n\n\n
features = self.extract_hu_moments(preprocessed)<\/p>\n\n\n\n
logging.info(“\u041e\u0431\u0440\u043e\u0431\u043b\u0435\u043d\u043e \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f: \u043e\u0442\u0440\u0438\u043c\u0430\u043d\u043e \u043e\u0437\u043d\u0430\u043a\u0438 %s”, features)<\/p>\n\n\n\n
return features<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 VoiceProcessor =======================<\/p>\n\n\n\n
class VoiceProcessor:<\/p>\n\n\n\n
def __init__(self, recognizer=None):<\/p>\n\n\n\n
self.recognizer = recognizer if recognizer is not None else sr.Recognizer()<\/p>\n\n\n\n
def process_audio(self, audio_path, engine=”sphinx”):<\/p>\n\n\n\n
with sr.AudioFile(audio_path) as source:<\/p>\n\n\n\n
audio = self.recognizer.record(source)<\/p>\n\n\n\n
try:<\/p>\n\n\n\n
if engine.lower() == “google”:<\/p>\n\n\n\n
text = self.recognizer.recognize_google(audio, language=”uk-UA”)<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
text = self.recognizer.recognize_sphinx(audio, language=”uk-UA”)<\/p>\n\n\n\n
logging.info(“\u0420\u043e\u0437\u043f\u0456\u0437\u043d\u0430\u043d\u043e \u0430\u0443\u0434\u0456\u043e: %s”, text)<\/p>\n\n\n\n
return text<\/p>\n\n\n\n
except sr.UnknownValueError:<\/p>\n\n\n\n
logging.warning(“\u0420\u043e\u0437\u043f\u0456\u0437\u043d\u0430\u0432\u0430\u043d\u043d\u044f \u0430\u0443\u0434\u0456\u043e \u043d\u0435\u0432\u0434\u0430\u043b\u043e\u0441\u044f (UnknownValueError)”)<\/p>\n\n\n\n
return “\u041d\u0435\u0432\u0456\u0434\u043e\u043c\u043e”<\/p>\n\n\n\n
except sr.RequestError as e:<\/p>\n\n\n\n
err_msg = f”\u041f\u043e\u043c\u0438\u043b\u043a\u0430 \u0441\u0435\u0440\u0432\u0456\u0441\u0443 \u0440\u043e\u0437\u043f\u0456\u0437\u043d\u0430\u0432\u0430\u043d\u043d\u044f: {e}”<\/p>\n\n\n\n
logging.error(err_msg)<\/p>\n\n\n\n
return err_msg<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 VideoProcessor =======================<\/p>\n\n\n\n
class VideoProcessor:<\/p>\n\n\n\n
def __init__(self, image_processor):<\/p>\n\n\n\n
self.image_processor = image_processor<\/p>\n\n\n\n
self.frame_cache = {}<\/p>\n\n\n\n
def process_frame(self, frame, frame_index):<\/p>\n\n\n\n
if frame_index in self.frame_cache:<\/p>\n\n\n\n
return self.frame_cache[frame_index]<\/p>\n\n\n\n
# \u042f\u043a\u0449\u043e \u043e\u0431\u0440\u043e\u0431\u043a\u0430 \u043f\u043e\u0432\u043d\u043e\u043a\u043e\u043b\u0456\u0440\u043d\u0430, \u0432\u0438\u043a\u043e\u0440\u0438\u0441\u0442\u043e\u0432\u0443\u0454\u0442\u044c\u0441\u044f \u043e\u0440\u0438\u0433\u0456\u043d\u0430\u043b\u044c\u043d\u0435 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f<\/p>\n\n\n\n
features = self.image_processor.extract_color_features(frame) if self.image_processor.process_color \\<\/p>\n\n\n\n
else self.image_processor.extract_hu_moments(frame)<\/p>\n\n\n\n
self.frame_cache[frame_index] = features<\/p>\n\n\n\n
return features<\/p>\n\n\n\n
def process_video(self, video_path, frame_interval=0.33):<\/p>\n\n\n\n
cap = cv2.VideoCapture(video_path)<\/p>\n\n\n\n
if not cap.isOpened():<\/p>\n\n\n\n
error_msg = “\u041d\u0435 \u0432\u0434\u0430\u043b\u043e\u0441\u044f \u0432\u0456\u0434\u043a\u0440\u0438\u0442\u0438 \u0432\u0456\u0434\u0435\u043e \u0437\u0430 \u0448\u043b\u044f\u0445\u043e\u043c: ” + video_path<\/p>\n\n\n\n
logging.error(error_msg)<\/p>\n\n\n\n
raise ValueError(error_msg)<\/p>\n\n\n\n
fps = cap.get(cv2.CAP_PROP_FPS)<\/p>\n\n\n\n
skip_frames = int(round(fps * frame_interval))<\/p>\n\n\n\n
features_list = []<\/p>\n\n\n\n
frame_index = 0<\/p>\n\n\n\n
tasks = []<\/p>\n\n\n\n
executor = ThreadPoolExecutor(max_workers=4)<\/p>\n\n\n\n
while cap.isOpened():<\/p>\n\n\n\n
ret, frame = cap.read()<\/p>\n\n\n\n
if not ret:<\/p>\n\n\n\n
break<\/p>\n\n\n\n
if frame_index % skip_frames == 0:<\/p>\n\n\n\n
tasks.append(executor.submit(self.process_frame, frame, frame_index))<\/p>\n\n\n\n
frame_index += 1<\/p>\n\n\n\n
cap.release()<\/p>\n\n\n\n
for future in as_completed(tasks):<\/p>\n\n\n\n
try:<\/p>\n\n\n\n
features_list.append(future.result())<\/p>\n\n\n\n
except Exception as exc:<\/p>\n\n\n\n
logging.error(“\u041f\u043e\u043c\u0438\u043b\u043a\u0430 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u043a\u0430\u0434\u0440\u0443: %s”, exc)<\/p>\n\n\n\n
executor.shutdown()<\/p>\n\n\n\n
logging.info(“\u041e\u0431\u0440\u043e\u0431\u043b\u0435\u043d\u043e \u0432\u0456\u0434\u0435\u043e: \u043e\u0442\u0440\u0438\u043c\u0430\u043d\u043e %s \u043d\u0430\u0431\u043e\u0440\u0456\u0432 \u043e\u0437\u043d\u0430\u043a”, len(features_list))<\/p>\n\n\n\n
return features_list<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 ImageGenerator =======================<\/p>\n\n\n\n
class ImageGenerator:<\/p>\n\n\n\n
def generate_image(self, query):<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
\u0413\u0435\u043d\u0435\u0440\u0443\u0454 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0430\u0431\u043e \u0434\u0456\u0430\u0433\u0440\u0430\u043c\u0443 \u0437\u0430 \u0437\u0430\u043f\u0438\u0442\u043e\u043c.<\/p>\n\n\n\n
\u042f\u043a\u0449\u043e \u0437\u0430\u043f\u0438\u0442 \u043c\u0456\u0441\u0442\u0438\u0442\u044c \u0441\u043b\u043e\u0432\u043e “diagram”, \u043c\u0430\u043b\u044e\u0454 \u043f\u0440\u043e\u0441\u0442\u0443 \u0441\u0445\u0435\u043c\u0443,<\/p>\n\n\n\n
\u0456\u043d\u0430\u043a\u0448\u0435 \u0441\u0442\u0432\u043e\u0440\u044e\u0454 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0437 \u0442\u0435\u043a\u0441\u0442\u043e\u043c.<\/p>\n\n\n\n
“””<\/p>\n\n\n\n
# \u0421\u0442\u0432\u043e\u0440\u0438\u043c\u043e \u043f\u043e\u0440\u043e\u0436\u043d\u0454 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f (500×500) \u0437 \u0431\u0456\u043b\u0438\u043c \u0444\u043e\u043d\u043e\u043c<\/p>\n\n\n\n
img = np.ones((500, 500, 3), dtype=np.uint8) * 255<\/p>\n\n\n\n
if “diagram” in query.lower():<\/p>\n\n\n\n
# \u041d\u0430\u043c\u0430\u043b\u044e\u0454\u043c\u043e \u043f\u0440\u043e\u0441\u0442\u0443 \u0434\u0456\u0430\u0433\u0440\u0430\u043c\u0443: \u043f\u0440\u044f\u043c\u043e\u043a\u0443\u0442\u043d\u0438\u043a, \u043a\u043e\u043b\u043e, \u043b\u0456\u043d\u0456\u044e \u0456 \u0442\u0435\u043a\u0441\u0442<\/p>\n\n\n\n
cv2.rectangle(img, (50, 50), (150, 150), (0, 0, 255), 3)<\/p>\n\n\n\n
cv2.circle(img, (300, 300), 50, (255, 0, 0), -1)<\/p>\n\n\n\n
cv2.line(img, (0, 0), (500, 500), (0, 255, 0), 2)<\/p>\n\n\n\n
cv2.putText(img, “Diagram”, (180, 250), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 2)<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
cv2.putText(img, query, (50, 250), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 2)<\/p>\n\n\n\n
logging.info(“\u0417\u0433\u0435\u043d\u0435\u0440\u043e\u0432\u0430\u043d\u043e \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0437\u0430 \u0437\u0430\u043f\u0438\u0442\u043e\u043c: %s”, query)<\/p>\n\n\n\n
return img<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 MemoryGraph =======================<\/p>\n\n\n\n
class MemoryGraph:<\/p>\n\n\n\n
def __init__(self):<\/p>\n\n\n\n
self.nodes = {}<\/p>\n\n\n\n
def add_node(self, node_id, data):<\/p>\n\n\n\n
self.nodes[node_id] = {“data”: data, “neighbors”: {}}<\/p>\n\n\n\n
logging.info(“\u0414\u043e\u0434\u0430\u043d\u043e \u0432\u0443\u0437\u043e\u043b \u0443 \u0433\u0440\u0430\u0444 \u043f\u0430\u043c’\u044f\u0442\u0456: %s”, node_id)<\/p>\n\n\n\n
def add_edge(self, node_id1, node_id2, weight=1.0):<\/p>\n\n\n\n
if node_id1 in self.nodes and node_id2 in self.nodes:<\/p>\n\n\n\n
self.nodes[node_id1][“neighbors”][node_id2] = weight<\/p>\n\n\n\n
self.nodes[node_id2][“neighbors”][node_id1] = weight<\/p>\n\n\n\n
logging.info(“\u0414\u043e\u0434\u0430\u043d\u043e \u0440\u0435\u0431\u0440\u043e \u043c\u0456\u0436 %s \u0442\u0430 %s”, node_id1, node_id2)<\/p>\n\n\n\n
def get_node(self, node_id):<\/p>\n\n\n\n
return self.nodes.get(node_id, None)<\/p>\n\n\n\n
def search_by_data(self, query):<\/p>\n\n\n\n
results = []<\/p>\n\n\n\n
for node_id, info in self.nodes.items():<\/p>\n\n\n\n
if query.lower() in str(info[“data”]).lower():<\/p>\n\n\n\n
results.append(node_id)<\/p>\n\n\n\n
logging.info(“\u041f\u043e\u0448\u0443\u043a \u0437\u0430 \u0434\u0430\u043d\u0438\u043c\u0438 ‘%s’ \u043f\u043e\u0432\u0435\u0440\u043d\u0443\u0432: %s”, query, results)<\/p>\n\n\n\n
return results<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 CognitiveMonitor =======================<\/p>\n\n\n\n
class CognitiveMonitor:<\/p>\n\n\n\n
def __init__(self, perceptron):<\/p>\n\n\n\n
self.perceptron = perceptron<\/p>\n\n\n\n
def report(self):<\/p>\n\n\n\n
states = self.perceptron.assess_network_state()<\/p>\n\n\n\n
avg_consciousness = sum(state[“consciousness_frequency”] for state in states) \/ len(states)<\/p>\n\n\n\n
logging.info(“Cognitive Monitor Report: Average Consciousness Frequency: %s”, avg_consciousness)<\/p>\n\n\n\n
def adjust_parameters(self, target_frequency=1.0):<\/p>\n\n\n\n
for neuron in self.perceptron.neurons:<\/p>\n\n\n\n
if neuron.consciousness_frequency < target_frequency:<\/p>\n\n\n\n
neuron.update_consciousness_parameters(consciousness_frequency=target_frequency)<\/p>\n\n\n\n
logging.info(“\u041d\u0430\u043b\u0430\u0448\u0442\u043e\u0432\u0430\u043d\u043e consciousness_frequency \u0434\u043b\u044f \u043d\u0435\u0439\u0440\u043e\u043d\u0430 \u0434\u043e %s”, target_frequency)<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 MultiModalProcessor =======================<\/p>\n\n\n\n
class MultiModalProcessor:<\/p>\n\n\n\n
def __init__(self, image_processor, voice_processor):<\/p>\n\n\n\n
self.image_processor = image_processor<\/p>\n\n\n\n
self.voice_processor = voice_processor<\/p>\n\n\n\n
def process(self, image_path=None, audio_path=None, text_input=None):<\/p>\n\n\n\n
results = {}<\/p>\n\n\n\n
if image_path:<\/p>\n\n\n\n
try:<\/p>\n\n\n\n
image_features = self.image_processor.process_image(image_path)<\/p>\n\n\n\n
results[“image_features”] = image_features<\/p>\n\n\n\n
except Exception as e:<\/p>\n\n\n\n
results[“image_features_error”] = str(e)<\/p>\n\n\n\n
logging.error(“\u041f\u043e\u043c\u0438\u043b\u043a\u0430 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f: %s”, e)<\/p>\n\n\n\n
if audio_path:<\/p>\n\n\n\n
voice_text = self.voice_processor.process_audio(audio_path, engine=”sphinx”)<\/p>\n\n\n\n
results[“voice_text”] = voice_text<\/p>\n\n\n\n
if text_input:<\/p>\n\n\n\n
results[“text_input”] = text_input<\/p>\n\n\n\n
logging.info(“\u0420\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0438 \u043c\u0443\u043b\u044c\u0442\u0438\u043c\u043e\u0434\u0430\u043b\u044c\u043d\u043e\u0457 \u043e\u0431\u0440\u043e\u0431\u043a\u0438: %s”, results)<\/p>\n\n\n\n
return results<\/p>\n\n\n\n
# ======================= \u041a\u043b\u0430\u0441 CognitiveSystem =======================<\/p>\n\n\n\n
class CognitiveSystem:<\/p>\n\n\n\n
def __init__(self):<\/p>\n\n\n\n
self.perceptron = Perceptron()<\/p>\n\n\n\n
self.memory_graph = MemoryGraph()<\/p>\n\n\n\n
self.image_processor = ImageProcessor(process_color=True)<\/p>\n\n\n\n
self.voice_processor = VoiceProcessor()<\/p>\n\n\n\n
self.multi_modal_processor = MultiModalProcessor(self.image_processor, self.voice_processor)<\/p>\n\n\n\n
self.monitor = CognitiveMonitor(self.perceptron)<\/p>\n\n\n\n
self.video_processor = VideoProcessor(self.image_processor)<\/p>\n\n\n\n
self.image_generator = ImageGenerator()<\/p>\n\n\n\n
logging.info(“CognitiveSystem \u0456\u043d\u0456\u0446\u0456\u0430\u043b\u0456\u0437\u043e\u0432\u0430\u043d\u043e.”)<\/p>\n\n\n\n
def add_neuron(self, neuron):<\/p>\n\n\n\n
self.perceptron.add_neuron(neuron)<\/p>\n\n\n\n
def integrate_modalities(self, image_path=None, audio_path=None, text_input=None):<\/p>\n\n\n\n
results = self.multi_modal_processor.process(image_path, audio_path, text_input)<\/p>\n\n\n\n
if “image_features” in results:<\/p>\n\n\n\n
node_id = f”img_{time.time()}”<\/p>\n\n\n\n
self.memory_graph.add_node(node_id, data=[results[“image_features”]])<\/p>\n\n\n\n
if “voice_text” in results:<\/p>\n\n\n\n
node_id = f”voice_{time.time()}”<\/p>\n\n\n\n
self.memory_graph.add_node(node_id, data=results[“voice_text”])<\/p>\n\n\n\n
if “text_input” in results:<\/p>\n\n\n\n
node_id = f”text_{time.time()}”<\/p>\n\n\n\n
self.memory_graph.add_node(node_id, data=results[“text_input”])<\/p>\n\n\n\n
return results<\/p>\n\n\n\n
def integrate_video(self, video_path):<\/p>\n\n\n\n
video_features = self.video_processor.process_video(video_path, frame_interval=0.33)<\/p>\n\n\n\n
if video_features:<\/p>\n\n\n\n
avg_features = np.mean(video_features, axis=0)<\/p>\n\n\n\n
node_id = f”video_{time.time()}”<\/p>\n\n\n\n
self.memory_graph.add_node(node_id, data=avg_features.tolist())<\/p>\n\n\n\n
logging.info(“\u0410\u0433\u0440\u0435\u0433\u043e\u0432\u0430\u043d\u0456 \u043e\u0437\u043d\u0430\u043a\u0438 \u0432\u0456\u0434\u0435\u043e: %s”, avg_features)<\/p>\n\n\n\n
return avg_features<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
return None<\/p>\n\n\n\n
def run(self, input_vector):<\/p>\n\n\n\n
activation = self.perceptron.evaluate_network(input_vector)<\/p>\n\n\n\n
logging.info(“Global Activation: %s”, activation)<\/p>\n\n\n\n
return activation<\/p>\n\n\n\n
def monitor_system(self):<\/p>\n\n\n\n
self.monitor.report()<\/p>\n\n\n\n
self.monitor.adjust_parameters(target_frequency=1.0)<\/p>\n\n\n\n
def generate_image(self, query):<\/p>\n\n\n\n
generated = self.image_generator.generate_image(query)<\/p>\n\n\n\n
node_id = f”gen_{time.time()}”<\/p>\n\n\n\n
# \u0417\u0431\u0435\u0440\u0456\u0433\u0430\u0454\u043c\u043e \u0437\u0433\u0435\u043d\u0435\u0440\u043e\u0432\u0430\u043d\u0435 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f (\u043c\u043e\u0436\u043d\u0430 \u043a\u043e\u043d\u0432\u0435\u0440\u0442\u0443\u0432\u0430\u0442\u0438 \u0443 \u0441\u043f\u0438\u0441\u043e\u043a \u043f\u0456\u043a\u0441\u0435\u043b\u0456\u0432)<\/p>\n\n\n\n
self.memory_graph.add_node(node_id, data=generated.tolist())<\/p>\n\n\n\n
logging.info(“\u0417\u0433\u0435\u043d\u0435\u0440\u043e\u0432\u0430\u043d\u043e \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0437\u0430 \u0437\u0430\u043f\u0438\u0442\u043e\u043c ‘%s’, \u0437\u0431\u0435\u0440\u0435\u0436\u0435\u043d\u043e \u044f\u043a %s”, query, node_id)<\/p>\n\n\n\n
return generated<\/p>\n\n\n\n
# ======================= \u041f\u0440\u0438\u043a\u043b\u0430\u0434 \u0432\u0438\u043a\u043e\u0440\u0438\u0441\u0442\u0430\u043d\u043d\u044f =======================<\/p>\n\n\n\n
if __name__ == “__main__”:<\/p>\n\n\n\n
# \u0421\u0442\u0432\u043e\u0440\u0435\u043d\u043d\u044f \u043a\u043e\u0433\u043d\u0456\u0442\u0438\u0432\u043d\u043e\u0457 \u0441\u0438\u0441\u0442\u0435\u043c\u0438<\/p>\n\n\n\n
system = CognitiveSystem()<\/p>\n\n\n\n
# \u0421\u0442\u0432\u043e\u0440\u0435\u043d\u043d\u044f \u0442\u0430 \u0434\u043e\u0434\u0430\u0432\u0430\u043d\u043d\u044f \u043d\u0435\u0439\u0440\u043e\u043d\u0456\u0432 \u0434\u043e \u0441\u0438\u0441\u0442\u0435\u043c\u0438<\/p>\n\n\n\n
neuron1 = Neuron(vector=[0.1, 0.2, 0.7],<\/p>\n\n\n\n
memory_file=’neuron1_memory.pkl’,<\/p>\n\n\n\n
consciousness_frequency=0.9,<\/p>\n\n\n\n
thought_speed=1.2,<\/p>\n\n\n\n
info_quality=0.9)<\/p>\n\n\n\n
neuron2 = Neuron(vector=[0.5, 0.4, 0.1],<\/p>\n\n\n\n
memory_file=’neuron2_memory.pkl’,<\/p>\n\n\n\n
consciousness_frequency=0.8,<\/p>\n\n\n\n
thought_speed=1.0,<\/p>\n\n\n\n
info_quality=1.0)<\/p>\n\n\n\n
neuron1.add_connection(neuron2, weight_coefficient=0.8)<\/p>\n\n\n\n
system.add_neuron(neuron1)<\/p>\n\n\n\n
system.add_neuron(neuron2)<\/p>\n\n\n\n
# \u0406\u043d\u0442\u0435\u0433\u0440\u0430\u0446\u0456\u044f \u043c\u0443\u043b\u044c\u0442\u0438\u043c\u043e\u0434\u0430\u043b\u044c\u043d\u0438\u0445 \u0434\u0430\u043d\u0438\u0445 (\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f, \u0430\u0443\u0434\u0456\u043e, \u0442\u0435\u043a\u0441\u0442)<\/p>\n\n\n\n
image_path = “sample_image.jpg” # \u0428\u043b\u044f\u0445 \u0434\u043e \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f<\/p>\n\n\n\n
audio_path = “sample_audio.wav” # \u0428\u043b\u044f\u0445 \u0434\u043e \u0430\u0443\u0434\u0456\u043e\u0444\u0430\u0439\u043b\u0443<\/p>\n\n\n\n
text_input = “\u0414\u043e\u0434\u0430\u0442\u043a\u043e\u0432\u0438\u0439 \u0442\u0435\u043a\u0441\u0442\u043e\u0432\u0438\u0439 \u0432\u0445\u0456\u0434”<\/p>\n\n\n\n
modalities = system.integrate_modalities(image_path=image_path,<\/p>\n\n\n\n
audio_path=audio_path,<\/p>\n\n\n\n
text_input=text_input)<\/p>\n\n\n\n
logging.info(“\u041c\u0443\u043b\u044c\u0442\u0438\u043c\u043e\u0434\u0430\u043b\u044c\u043d\u0456 \u0434\u0430\u043d\u0456: %s”, modalities)<\/p>\n\n\n\n
# \u0406\u043d\u0442\u0435\u0433\u0440\u0430\u0446\u0456\u044f \u0432\u0456\u0434\u0435\u043e (\u043a\u0430\u0434\u0440\u0438 \u043a\u043e\u0436\u043d\u0456 0.33 \u0441\u0435\u043a\u0443\u043d\u0434\u0438)<\/p>\n\n\n\n
video_path = “sample_video.mp4” # \u0428\u043b\u044f\u0445 \u0434\u043e \u0432\u0456\u0434\u0435\u043e\u0444\u0430\u0439\u043b\u0443<\/p>\n\n\n\n
try:<\/p>\n\n\n\n
video_features = system.integrate_video(video_path)<\/p>\n\n\n\n
logging.info(“\u0410\u0433\u0440\u0435\u0433\u043e\u0432\u0430\u043d\u0456 \u043e\u0437\u043d\u0430\u043a\u0438 \u0432\u0456\u0434\u0435\u043e: %s”, video_features)<\/p>\n\n\n\n
except Exception as e:<\/p>\n\n\n\n
logging.error(“\u041f\u043e\u043c\u0438\u043b\u043a\u0430 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u0432\u0456\u0434\u0435\u043e: %s”, e)<\/p>\n\n\n\n
# \u0412\u0438\u043a\u043e\u0440\u0438\u0441\u0442\u0430\u043d\u043d\u044f \u043e\u0442\u0440\u0438\u043c\u0430\u043d\u0438\u0445 \u043e\u0437\u043d\u0430\u043a \u044f\u043a \u0432\u0445\u0456\u0434\u043d\u043e\u0433\u043e \u0441\u0438\u0433\u043d\u0430\u043b\u0443<\/p>\n\n\n\n
if modalities.get(“image_features”) is not None:<\/p>\n\n\n\n
input_vector = modalities[“image_features”]<\/p>\n\n\n\n
elif video_features is not None:<\/p>\n\n\n\n
input_vector = video_features<\/p>\n\n\n\n
else:<\/p>\n\n\n\n
input_vector = [0.3, 0.6, 0.1]<\/p>\n\n\n\n
system.run(input_vector)<\/p>\n\n\n\n
# \u0413\u0435\u043d\u0435\u0440\u0430\u0446\u0456\u044f \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f (\u0434\u0456\u0430\u0433\u0440\u0430\u043c\u0438) \u0437\u0430 \u0437\u0430\u043f\u0438\u0442\u043e\u043c<\/p>\n\n\n\n
generated_img = system.generate_image(“Show me a diagram”)<\/p>\n\n\n\n
cv2.imwrite(“generated_diagram.jpg”, generated_img)<\/p>\n\n\n\n
logging.info(“\u0417\u0433\u0435\u043d\u0435\u0440\u043e\u0432\u0430\u043d\u0435 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0437\u0431\u0435\u0440\u0435\u0436\u0435\u043d\u043e \u044f\u043a generated_diagram.jpg”)<\/p>\n\n\n\n
# \u0412\u0438\u043a\u043e\u043d\u0430\u043d\u043d\u044f \u043c\u0435\u0442\u0430\u043a\u043e\u0433\u043d\u0456\u0442\u0438\u0432\u043d\u043e\u0433\u043e \u043c\u043e\u043d\u0456\u0442\u043e\u0440\u0438\u043d\u0433\u0443<\/p>\n\n\n\n
system.monitor_system()<\/p>\n\n\n\n
# \u041f\u043e\u0431\u0443\u0434\u043e\u0432\u0430 \u043c\u0435\u043d\u0442\u0430\u043b\u044c\u043d\u043e\u0457 \u043a\u043e\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0456\u0457 \u043d\u0430 \u043e\u0441\u043d\u043e\u0432\u0456 \u043b\u043e\u043a\u0443\u0441\u0456\u0432 \u043f\u0430\u043c’\u044f\u0442\u0456 \u0437 \u043d\u0435\u0439\u0440\u043e\u043d\u0456\u0432<\/p>\n\n\n\n
neuron1.store_memory(“A”, “\u041f\u043e\u043d\u044f\u0442\u0442\u044f A”, weight=0.9, category=”fact”)<\/p>\n\n\n\n
neuron1.store_memory(“B”, “\u041f\u043e\u043d\u044f\u0442\u0442\u044f B”, weight=0.8, category=”a_priori”)<\/p>\n\n\n\n
construction = system.perceptron.build_mental_construction(“Construction1”, [“A”, “B”])<\/p>\n\n\n\n
logging.info(“\u041c\u0435\u043d\u0442\u0430\u043b\u044c\u043d\u0430 \u043a\u043e\u043d\u0441\u0442\u0440\u0443\u043a\u0446\u0456\u044f ‘Construction1’: %s”, construction)<\/p>\n\n\n\n
# \u0412\u0438\u0432\u0456\u0434 \u0433\u0440\u0430\u0444\u0443 \u043f\u0430\u043c’\u044f\u0442\u0456 (\u0441\u043f\u0438\u0441\u043e\u043a \u0432\u0443\u0437\u043b\u0456\u0432)<\/p>\n\n\n\n
logging.info(“\u0412\u0443\u0437\u043b\u0438 \u0433\u0440\u0430\u0444\u0443 \u043f\u0430\u043c’\u044f\u0442\u0456:”)<\/p>\n\n\n\n
for node_id, node_info in system.memory_graph.nodes.items():<\/p>\n\n\n\n
logging.info(“%s: %s”, node_id, node_info)<\/p>\n\n\n\n
\u041e\u043f\u0438\u0441 \u0441\u0438\u0441\u0442\u0435\u043c\u0438 \u043f\u0440\u043e\u0441\u0442\u0438\u043c\u0438 \u0441\u043b\u043e\u0432\u0430\u043c\u0438<\/strong><\/p>\n\n\n\n \u0426\u0435\u0439 \u043a\u043e\u0434 \u0441\u0442\u0432\u043e\u0440\u044e\u0454 \u043a\u043e\u043c\u043f\u043b\u0435\u043a\u0441\u043d\u0443 \u043a\u043e\u0433\u043d\u0456\u0442\u0438\u0432\u043d\u0443 \u0441\u0438\u0441\u0442\u0435\u043c\u0443, \u044f\u043a\u0430 \u0437\u0434\u0430\u0442\u043d\u0430:<\/p>\n\n\n\n \u0422\u0430\u043a\u0438\u043c \u0447\u0438\u043d\u043e\u043c, \u0441\u0438\u0441\u0442\u0435\u043c\u0430 \u043c\u043e\u0436\u0435 \u043d\u0435 \u043b\u0438\u0448\u0435 \u043f\u0440\u0438\u0439\u043c\u0430\u0442\u0438 \u0442\u0430 \u0430\u043d\u0430\u043b\u0456\u0437\u0443\u0432\u0430\u0442\u0438 \u0440\u0456\u0437\u043d\u043e\u043c\u043e\u0434\u0430\u043b\u044c\u043d\u0456 \u0434\u0430\u043d\u0456, \u0430 \u0439 \u0441\u0430\u043c\u0430 \u0441\u0442\u0432\u043e\u0440\u044e\u0432\u0430\u0442\u0438 \u0432\u0456\u0437\u0443\u0430\u043b\u044c\u043d\u0438\u0439 \u043a\u043e\u043d\u0442\u0435\u043d\u0442 (\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f, \u0441\u0445\u0435\u043c\u0438) \u0437\u0430 \u043f\u043e\u0442\u0440\u0435\u0431\u043e\u044e. \u0426\u0435 \u0440\u043e\u0431\u0438\u0442\u044c \u0457\u0457 \u043f\u043e\u0442\u0443\u0436\u043d\u0438\u043c \u0456\u043d\u0441\u0442\u0440\u0443\u043c\u0435\u043d\u0442\u043e\u043c \u0434\u043b\u044f \u043c\u043e\u0434\u0435\u043b\u044e\u0432\u0430\u043d\u043d\u044f \u043a\u043e\u0433\u043d\u0456\u0442\u0438\u0432\u043d\u0438\u0445 \u043f\u0440\u043e\u0446\u0435\u0441\u0456\u0432 \u0456 \u0434\u043b\u044f \u0437\u0430\u0441\u0442\u043e\u0441\u0443\u0432\u0430\u043d\u043d\u044f \u0432 \u0456\u043d\u0442\u0435\u0440\u0430\u043a\u0442\u0438\u0432\u043d\u0438\u0445 \u0430\u0441\u0438\u0441\u0442\u0435\u043d\u0442\u0430\u0445, \u0430\u043d\u0430\u043b\u0456\u0437\u0456 \u043c\u0435\u0434\u0456\u0430 \u0430\u0431\u043e \u043d\u0430\u0432\u0456\u0442\u044c \u0433\u0435\u043d\u0435\u0440\u0430\u0446\u0456\u0457 \u0432\u0456\u0437\u0443\u0430\u043b\u044c\u043d\u0438\u0445 \u043f\u043e\u044f\u0441\u043d\u0435\u043d\u044c.<\/p>\n","protected":false},"excerpt":{"rendered":" \u0410\u0432\u0442\u043e\u0440 – \u0421\u0443\u0445\u0430\u0447\u043e\u0432 \u0414\u0435\u043d\u0438\u0441 \u041f\u0430\u0432\u043b\u043e\u0432\u0438\u0447 This is a work, and the author’s right to a work under international law comes into force from the moment the work is created. \u041d\u0438\u0436\u0447\u0435 \u043d\u0430\u0432\u0435\u0434\u0435\u043d\u043e \u043e\u043d\u043e\u0432\u043b\u0435\u043d\u0438\u0439 \u0432\u0430\u0440\u0456\u0430\u043d\u0442 \u043a\u043e\u0434\u0443 \u0437 \u043d\u0430\u0441\u0442\u0443\u043f\u043d\u0438\u043c\u0438 \u0434\u043e\u043f\u043e\u0432\u043d\u0435\u043d\u043d\u044f\u043c\u0438: python \u041a\u043e\u043f\u0456\u044e\u0432\u0430\u0442\u0438 import numpy as np import pickle import os import time import cv2 import logging import speech_recognition as sr from concurrent.futures import ThreadPoolExecutor, as_completed # \u041d\u0430\u043b\u0430\u0448\u0442\u0443\u0432\u0430\u043d\u043d\u044f \u043b\u043e\u0433\u0443\u0432\u0430\u043d\u043d\u044f logging.basicConfig(level=logging.INFO, format=’%(asctime)s [%(levelname)s] %(message)s’) # ======================= \u041a\u043b\u0430\u0441 Neuron ======================= class Neuron: def __init__(self, vector=None, […]<\/p>\n","protected":false},"author":1,"featured_media":1400,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1],"tags":[],"class_list":["post-1545","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-bez-kategoriyi"],"yoast_head":"\n\n
\u0417\u0430\u043c\u0456\u0441\u0442\u044c \u043f\u0435\u0440\u0435\u0442\u0432\u043e\u0440\u0435\u043d\u043d\u044f \u0432 \u0447\u043e\u0440\u043d\u043e-\u0431\u0456\u043b\u0438\u0439 \u0444\u043e\u0440\u043c\u0430\u0442, \u0441\u0438\u0441\u0442\u0435\u043c\u0430 \u043e\u0431\u0440\u043e\u0431\u043b\u044f\u0454 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0432 \u043a\u043e\u043b\u044c\u043e\u0440\u0456, \u0432\u0438\u0434\u0456\u043b\u044f\u044e\u0447\u0438 \u0441\u0435\u0440\u0435\u0434\u043d\u0456 \u0437\u043d\u0430\u0447\u0435\u043d\u043d\u044f \u0434\u043b\u044f \u043a\u043e\u0436\u043d\u043e\u0433\u043e \u0437 \u043a\u0430\u043d\u0430\u043b\u0456\u0432 (B, G, R). \u0426\u0435 \u0434\u043e\u0437\u0432\u043e\u043b\u044f\u0454 \u00ab\u0431\u0430\u0447\u0438\u0442\u0438\u00bb \u0432\u0441\u044e \u043a\u043e\u043b\u0456\u0440\u043d\u0443 \u043f\u0430\u043b\u0456\u0442\u0440\u0443.<\/li>\n\n\n\n
\u041c\u043e\u0434\u0443\u043b\u044c VoiceProcessor \u043f\u0435\u0440\u0435\u0442\u0432\u043e\u0440\u044e\u0454 \u0430\u0443\u0434\u0456\u043e\u0444\u0430\u0439\u043b\u0438 (\u0433\u043e\u043b\u043e\u0441) \u043d\u0430 \u0442\u0435\u043a\u0441\u0442, \u0432\u0438\u043a\u043e\u0440\u0438\u0441\u0442\u043e\u0432\u0443\u044e\u0447\u0438 \u0430\u043b\u0433\u043e\u0440\u0438\u0442\u043c\u0438 \u0440\u043e\u0437\u043f\u0456\u0437\u043d\u0430\u0432\u0430\u043d\u043d\u044f (\u043d\u0430\u043f\u0440\u0438\u043a\u043b\u0430\u0434, Sphinx \u0447\u0438 Google).<\/li>\n\n\n\n
\u041c\u043e\u0434\u0443\u043b\u044c VideoProcessor \u0432\u0438\u0431\u0438\u0440\u0430\u0454 \u043a\u0430\u0434\u0440\u0438 \u0437 \u0432\u0456\u0434\u0435\u043e \u043a\u043e\u0436\u043d\u0456 0.33 \u0441\u0435\u043a\u0443\u043d\u0434\u0438, \u043e\u0431\u0440\u043e\u0431\u043b\u044f\u0454 \u0457\u0445 \u043f\u0430\u0440\u0430\u043b\u0435\u043b\u044c\u043d\u043e, \u043e\u0442\u0440\u0438\u043c\u0443\u044e\u0447\u0438 \u043e\u0437\u043d\u0430\u043a\u0438 (\u043d\u0430\u043f\u0440\u0438\u043a\u043b\u0430\u0434, \u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u043a\u043e\u043b\u0456\u0440 \u043a\u043e\u0436\u043d\u043e\u0433\u043e \u043a\u0430\u0434\u0440\u0443). \u041e\u0442\u0440\u0438\u043c\u0430\u043d\u0456 \u043e\u0437\u043d\u0430\u043a\u0438 \u043c\u043e\u0436\u0443\u0442\u044c \u0431\u0443\u0442\u0438 \u0430\u0433\u0440\u0435\u0433\u043e\u0432\u0430\u043d\u0456 \u0434\u043b\u044f \u043e\u043f\u0438\u0441\u0443 \u0432\u0441\u044c\u043e\u0433\u043e \u0432\u0456\u0434\u0435\u043e.<\/li>\n\n\n\n
\u041c\u043e\u0434\u0443\u043b\u044c MemoryGraph \u0437\u0431\u0435\u0440\u0456\u0433\u0430\u0454 \u00ab\u043b\u043e\u043a\u0443\u0441\u0438\u00bb \u043f\u0430\u043c\u2019\u044f\u0442\u0456 (\u043d\u0430\u043f\u0440\u0438\u043a\u043b\u0430\u0434, \u0440\u0435\u0437\u0443\u043b\u044c\u0442\u0430\u0442\u0438 \u043e\u0431\u0440\u043e\u0431\u043a\u0438 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u044c, \u0430\u0443\u0434\u0456\u043e, \u0432\u0456\u0434\u0435\u043e \u0430\u0431\u043e \u0437\u0433\u0435\u043d\u0435\u0440\u043e\u0432\u0430\u043d\u0456 \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f) \u0456 \u0434\u043e\u0437\u0432\u043e\u043b\u044f\u0454 \u0432\u0441\u0442\u0430\u043d\u043e\u0432\u043b\u044e\u0432\u0430\u0442\u0438 \u0430\u0441\u043e\u0446\u0456\u0430\u0446\u0456\u0457 \u043c\u0456\u0436 \u043d\u0438\u043c\u0438.<\/li>\n\n\n\n
CognitiveMonitor \u0430\u043d\u0430\u043b\u0456\u0437\u0443\u0454 \u0441\u0442\u0430\u043d \u043c\u0435\u0440\u0435\u0436\u0456 (\u0441\u0435\u0440\u0435\u0434\u043d\u0456\u0439 \u0440\u0456\u0432\u0435\u043d\u044c \u00ab\u0441\u0432\u0456\u0434\u043e\u043c\u043e\u0441\u0442\u0456\u00bb \u043d\u0435\u0439\u0440\u043e\u043d\u0456\u0432) \u0456 \u0437\u0430 \u043f\u043e\u0442\u0440\u0435\u0431\u0438 \u043a\u043e\u0440\u0438\u0433\u0443\u0454 \u043f\u0430\u0440\u0430\u043c\u0435\u0442\u0440\u0438, \u0449\u043e\u0431 \u0437\u0430\u0431\u0435\u0437\u043f\u0435\u0447\u0438\u0442\u0438 \u0441\u0442\u0430\u0431\u0456\u043b\u044c\u043d\u0443 \u0440\u043e\u0431\u043e\u0442\u0443 \u0441\u0438\u0441\u0442\u0435\u043c\u0438.<\/li>\n\n\n\n
MultiModalProcessor \u043e\u0431\u2019\u0454\u0434\u043d\u0443\u0454 \u0434\u0430\u043d\u0456 \u0456\u0437 \u0440\u0456\u0437\u043d\u0438\u0445 \u0441\u0435\u043d\u0441\u043e\u0440\u043d\u0438\u0445 \u043a\u0430\u043d\u0430\u043b\u0456\u0432 (\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f, \u0433\u043e\u043b\u043e\u0441, \u0442\u0435\u043a\u0441\u0442) \u0434\u043b\u044f \u043f\u043e\u0431\u0443\u0434\u043e\u0432\u0438 \u0454\u0434\u0438\u043d\u043e\u0457 \u043c\u043e\u0434\u0435\u043b\u0456 \u0437\u043d\u0430\u043d\u044c.<\/li>\n\n\n\n
ImageGenerator \u0434\u043e\u0437\u0432\u043e\u043b\u044f\u0454 \u0441\u0438\u0441\u0442\u0435\u043c\u0456 \u00ab\u043c\u0430\u043b\u044e\u0432\u0430\u0442\u0438\u00bb \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0430\u0431\u043e \u0434\u0456\u0430\u0433\u0440\u0430\u043c\u0438 \u043d\u0430 \u043e\u0441\u043d\u043e\u0432\u0456 \u0442\u0435\u043a\u0441\u0442\u043e\u0432\u043e\u0433\u043e \u0437\u0430\u043f\u0438\u0442\u0443. \u041d\u0430\u043f\u0440\u0438\u043a\u043b\u0430\u0434, \u0437\u0430 \u0437\u0430\u043f\u0438\u0442\u043e\u043c \u00abShow me a diagram\u00bb \u0441\u0438\u0441\u0442\u0435\u043c\u0430 \u0441\u0442\u0432\u043e\u0440\u0438\u0442\u044c \u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u043d\u044f \u0437 \u043f\u0440\u043e\u0441\u0442\u0438\u043c\u0438 \u0433\u0435\u043e\u043c\u0435\u0442\u0440\u0438\u0447\u043d\u0438\u043c\u0438 \u0444\u0456\u0433\u0443\u0440\u0430\u043c\u0438 \u0442\u0430 \u0442\u0435\u043a\u0441\u0442\u043e\u043c.<\/li>\n<\/ul>\n\n\n\n