Software engineer by trade,
AI enthusiast by obsession.

CODE SAMPLES

Python

# Loss and optimizer

criterion = nn.CrossEntropyLoss()

optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)

# The training loop

for epoch in range(num_epochs):

for (words, labels) in train_loader:

words = words.to(device)

labels = labels.to(dtype=torch.long).to(device)

# Forward pass

outputs = model(words)

loss = criterion(outputs, labels)

# Backward and optimize

optimizer.zero_grad()

loss.backward()

optimizer.step()

if (epoch+1) % 100 == 0:

print(f'Epoch [{epoch+1}/{num_epochs}], Loss: {loss.item():.4f}')

Python

# Set up SOCKS proxy

quotaguard_url = os.getenv('QUOTAGUARDSTATIC_SOCKS_URL')

if quotaguard_url:

parsed = urlparse(quotaguard_url)

socks.set_default_proxy(

socks.PROXY_TYPE_SOCKS5,

parsed.hostname,

parsed.port,

parsed.username,

parsed.password

)

socket.socket = socks.socksocket

Python

# Precompute intent embeddings

all_patterns = []

pattern_tags = []

# Load all patterns from intents file

for intent in intents['intents']:

tag = intent['tag']

for pattern in intent['patterns']:

cleaned_pattern = clean_sentence(pattern)

all_patterns.append(cleaned_pattern)

pattern_tags.append(tag)

pattern_embeddings = sentence_model.encode(all_patterns)

logger.debug(f"Loaded and pre-computed embeddings for {len(all_patterns)} patterns.")