Text Generation Models

Using Text Generation Models

from transformers import pipeline, AutoModelForCausalLM, AutoTokenizer
import torch

class TextGenerator:
    def __init__(self, model_name="gpt2"):
        self.tokenizer = AutoTokenizer.from_pretrained(model_name)
        self.model = AutoModelForCausalLM.from_pretrained(model_name)

    def generate(self, prompt, max_length=100, temperature=0.7,
                 top_k=50, top_p=0.95, do_sample=True):
        inputs = self.tokenizer.encode(prompt, return_tensors="pt")

        with torch.no_grad():
            outputs = self.model.generate(
                inputs,
                max_length=max_length,
                temperature=temperature,
                top_k=top_k,
                top_p=top_p,
                do_sample=do_sample,
                pad_token_id=self.tokenizer.eos_token_id
            )

        return self.tokenizer.decode(outputs[0], skip_special_tokens=True)

# Usage
generator = TextGenerator("gpt2")
text = generator.generate("The future of AI is")
print(text)

Decoding Strategies

Model Selection Guide

Summary

Text generation models vary in size, capability, and accessibility. Choose based on your specific requirements for quality, cost, and deployment constraints.

Next: We'll explore instruction tuning techniques.