Some Python and LLM related code snippets

1. Using LLMs to Generate Code

python
import openai

# Define the prompt
prompt = "Write a Python function that takes a list of numbers and returns the sum of the squares of the numbers."

# Use the OpenAI API to generate the code
response = openai.Completion.create(
    model="text-davinci-002",
    prompt=prompt,
    max_tokens=100,
    temperature=0.5,
    stop=["\n"]
)

# Print the generated code
print(response.choices[0].text)

2. Fine-tuning LLMs for Code Generation

python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load the pre-trained model and tokenizer
model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-medium")
tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-medium")

# Define the fine-tuning dataset
train_dataset = ...

# Fine-tune the model
model.train()
for epoch in range(5):
    for batch in train_dataset:
        inputs = tokenizer(batch, return_tensors="pt", max_length=512, padding="max_length", truncation=True)
        labels = tokenizer(batch, return_tensors="pt", max_length=512, padding="max_length", truncation=True)
        model.zero_grad()
        outputs = model(**inputs, labels=labels)
        loss = outputs.loss
        loss.backward()
        model.optimizers[0].step()

# Use the fine-tuned model for code generation
prompt = "Write a Python function that takes a list of numbers and returns the sum of the squares of the numbers."
input_ids = tokenizer(prompt, return_tensors="pt", max_length=512, padding="max_length", truncation=True)
output = model.generate(input_ids, max_length=100, temperature=0.5)
print(tokenizer.decode(output[0], skip_special_tokens=True))

3. Using LLMs as Virtual Assistants

python
import openai

# Define the prompt
prompt = "Write a Python function that takes a list of numbers and returns the sum of the squares of the numbers."

# Use the OpenAI API to generate the code
response = openai.Completion.create(
    model="text-davinci-002",
    prompt=prompt,
    max_tokens=100,
    temperature=0.5,
    stop=["\n"]
)

# Print the generated code
print(response.choices[0].text)

# Use the generated code as a virtual assistant
def sum_of_squares(numbers):
    return sum([num ** 2 for num in numbers])

# Test the virtual assistant
numbers = [1, 2, 3, 4, 5]
result = sum_of_squares(numbers)
print(result)

4. Automating Code Refinement

python
import openai

# Define the prompt
prompt = "Refine the following Python code to handle edge cases and improve performance."

# Use the OpenAI API to generate the refined code
response = openai.Completion.create(
    model="text-davinci-002",
    prompt=prompt,
    max_tokens=100,
    temperature=0.5,
    stop=["\n"]
)

# Print the refined code
print(response.choices[0].text)

# Use the refined code as a virtual assistant
# ...

# Test the refined code
# ...