| |
| |
| |
| import os |
| import torch |
| from transformers import AutoModelForCausalLM, AutoTokenizer |
|
|
| |
| |
| MODEL_NAME = "meta-llama/Llama-2-7b-chat-hf" |
| DEVICE = torch.device("cuda" if torch.cuda.is_available() else "cpu") |
|
|
| |
| print(f"🔄 تحميل النموذج {MODEL_NAME} على {DEVICE} …") |
| tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) |
| model = AutoModelForCausalLM.from_pretrained( |
| MODEL_NAME, |
| torch_dtype=torch.float16 if DEVICE.type == "cuda" else torch.float32, |
| low_cpu_mem_usage=True |
| ).to(DEVICE) |
|
|
| print("✅ النموذج جاهز!\n") |
|
|
| |
| def generate(prompt: str, max_new_tokens: int = 150, temperature: float = 0.7): |
| inputs = tokenizer(prompt, return_tensors="pt").to(DEVICE) |
|
|
| with torch.no_grad(): |
| output = model.generate( |
| **inputs, |
| max_new_tokens=max_new_tokens, |
| do_sample=True, |
| temperature=temperature, |
| top_p=0.9, |
| pad_token_id=tokenizer.eos_token_id, |
| ) |
| |
| text = tokenizer.decode(output[0], skip_special_tokens=True) |
| return text[len(prompt):].strip() |
|
|
| if __name__ == "__main__": |
| print("🗨️ اكتب جملة (أو اكتب 'exit' للخروج)") |
| while True: |
| user_input = input("\n>>> ") |
| if user_input.lower() in {"exit", "quit"}: |
| print("👋 وداعًا!") |
| break |
|
|
| answer = generate(user_input) |
| print(f"\n🤖 {answer}") |
