Spaces:
Runtime error
Runtime error
Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
| 3 |
+
import torch
|
| 4 |
+
|
| 5 |
+
# Načtení modelu
|
| 6 |
+
MODEL_NAME = "meta-llama/Llama-2-7b-chat-hf" # Změň podle potřeby
|
| 7 |
+
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
|
| 8 |
+
model = AutoModelForCausalLM.from_pretrained(
|
| 9 |
+
MODEL_NAME,
|
| 10 |
+
torch_dtype=torch.float16,
|
| 11 |
+
device_map="auto"
|
| 12 |
+
)
|
| 13 |
+
|
| 14 |
+
# Funkce pro generování textu
|
| 15 |
+
def chat(prompt):
|
| 16 |
+
inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
|
| 17 |
+
output = model.generate(**inputs, max_new_tokens=100)
|
| 18 |
+
return tokenizer.decode(output[0], skip_special_tokens=True)
|
| 19 |
+
|
| 20 |
+
# Webové UI pomocí Gradio
|
| 21 |
+
interface = gr.Interface(
|
| 22 |
+
fn=chat,
|
| 23 |
+
inputs="text",
|
| 24 |
+
outputs="text",
|
| 25 |
+
title="LLaMA 2 Chatbot",
|
| 26 |
+
description="Tento chatbot běží na modelu LLaMA 2 a odpovídá na otázky.",
|
| 27 |
+
)
|
| 28 |
+
|
| 29 |
+
interface.launch()
|