| from fastapi import FastAPI, Request, Form |
| from fastapi.responses import HTMLResponse |
| from transformers import GPT2LMHeadModel, GPT2Config |
| import torch |
|
|
| app = FastAPI() |
|
|
| quantized_model_path = "gpt3_mini_quantized_2x_16bits.pth" |
| config = GPT2Config.from_pretrained("Deniskin/gpt3_medium") |
| quantized_model = GPT2LMHeadModel(config=config) |
|
|
| |
| model.eval() |
|
|
| |
| def generate_text(prompt): |
| input_ids = tokenizer.encode(prompt, return_tensors="pt") |
| output = model.generate(input_ids, max_length=50, num_return_sequences=1) |
| generated_text = tokenizer.decode(output[0], skip_special_tokens=True) |
| return generated_text |
|
|
| @app.get("/", response_class=HTMLResponse) |
| async def home(request: Request): |
| html_content = """ |
| <!DOCTYPE html> |
| <html lang="en"> |
| <head> |
| <meta charset="UTF-8"> |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> |
| <title>LSTM Text Generation</title> |
| </head> |
| <body> |
| <h1>LSTM Text Generation</h1> |
| <form id="text-form"> |
| <label for="user-input">Enter your input:</label><br> |
| <textarea id="user-input" name="user-input" rows="4" cols="50"></textarea><br> |
| <button type="submit">Generate Text</button> |
| </form> |
| <div id="output"></div> |
| |
| <script> |
| document.getElementById("text-form").addEventListener("submit", function(event) { |
| event.preventDefault(); |
| var userInput = document.getElementById("user-input").value; |
| |
| fetch("/generate", { |
| method: "POST", |
| headers: { |
| "Content-Type": "application/json" |
| }, |
| body: JSON.stringify({ input_text: userInput }) |
| }) |
| .then(response => response.json()) |
| .then(data => { |
| document.getElementById("output").innerText = data.generated_text; |
| }); |
| }); |
| </script> |
| </body> |
| </html> |
| """ |
| return HTMLResponse(content=html_content, status_code=200) |
|
|
| @app.post("/generate") |
| async def generate(request: Request, input_text: str = Form(...)): |
| generated_text = generate_text(input_text) |
| return {"generated_text": generated_text} |
|
|
| if __name__ == "__main__": |
| import uvicorn |
| uvicorn.run(app, host="0.0.0.0", port=8000) |
|
|
|
|
