Spaces:
Sleeping
Sleeping
File size: 1,499 Bytes
73bcc15 3a89aa0 73bcc15 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 | import sys
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
# -------------------------
# CONFIGURATION
# -------------------------
BASE_MODEL = "microsoft/phi-3-medium-4k-instruct"
LORA_MODEL = "KayHamla06/Milo_WBA" # Hugging Face repo ID
# -------------------------
# LOAD TOKENIZER + MODEL
# -------------------------
print("Loading tokenizer...")
tokenizer = AutoTokenizer.from_pretrained(BASE_MODEL)
print("Loading base model...")
model = AutoModelForCausalLM.from_pretrained(
BASE_MODEL,
torch_dtype=torch.float16,
device_map="auto"
)
print("Applying LoRA fine-tune...")
model = PeftModel.from_pretrained(model, LORA_MODEL)
model.eval()
# -------------------------
# INFERENCE FUNCTION
# -------------------------
def generate_response(prompt: str) -> str:
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
output_tokens = model.generate(
**inputs,
max_new_tokens=300,
temperature=0.7,
top_p=0.9,
do_sample=True,
pad_token_id=tokenizer.eos_token_id
)
return tokenizer.decode(output_tokens[0], skip_special_tokens=True)
# -------------------------
# RUN PROMPT FROM TERMINAL
# -------------------------
if __name__ == "__main__":
if len(sys.argv) < 2:
print("Usage: python model.py \"your prompt here\"")
sys.exit(1)
prompt = sys.argv[1]
print("\n--- MILO RESPONSE ---\n")
print(generate_response(prompt))
|