File size: 4,137 Bytes
51a7d9e bd34f0b 51a7d9e edb9e8a 51a7d9e c507b10 1ec2e60 51a7d9e 8830af9 51a7d9e bd34f0b 566d6bf bd34f0b 51a7d9e f1d6fbe 8830af9 51a7d9e 69855bc bd34f0b 51a7d9e 3f6e58a 51a7d9e bd34f0b fd6304d 51a7d9e 33e87c8 3b9cb87 bd34f0b 3b9cb87 bd34f0b 639e063 edb9e8a bd34f0b edb9e8a bd34f0b 51a7d9e ef2eb9e 51a7d9e edb9e8a 51a7d9e edb9e8a 51a7d9e a3e36c2 51a7d9e 86f332d 51a7d9e 579ca70 51a7d9e ef2eb9e 51a7d9e bd34f0b 51a7d9e 16e5a54 51a7d9e |
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 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 |
import torch
from PIL import Image
import gradio as gr
import spaces
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer
import os
from threading import Thread
HF_TOKEN = os.environ.get("HF_TOKEN", None)
MODEL_ID = "elyza/Llama-3-ELYZA-JP-8B"
MODELS = os.environ.get("MODELS")
MODEL_NAME = MODELS.split("/")[-1]
TITLE = "<h1><center>Llama-3-ELYZA-JP-8B Chat webui</center></h1>"
DESCRIPTION = f"""
<h3>MODEL: <a href="https://hf.co/{MODELS}">{MODEL_NAME}</a></h3>
<center>
<p>youko-8b is the large language model built by rinna.
<br>
Feel free to test without log.
</p>
</center>
"""
model = AutoModelForCausalLM.from_pretrained(
MODEL_ID,
torch_dtype=torch.float16,
device_map="auto",
)
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
@spaces.GPU
def stream_chat(message: str, history: list, temperature: float, max_new_tokens: int, top_p: float, top_k: int, penalty: float):
print(f'message is - {message}')
print(f'history is - {history}')
conversation = []
for prompt, answer in history:
conversation.extend([{"role": "user", "content": prompt}, {"role": "assistant", "content": answer}])
conversation.append({"role": "user", "content": message})
#print(f"Conversation is -\n{conversation}")
input_ids = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(input_ids, return_tensors="pt").to(0)
streamer = TextIteratorStreamer(tokenizer, timeout=10., skip_prompt=True, skip_special_tokens=True)
generate_kwargs = dict(
inputs,
streamer=streamer,
top_k=top_k,
top_p=top_p,
repetition_penalty=penalty,
max_new_tokens=max_new_tokens,
do_sample=True,
temperature=temperature,
eos_token_id = [128001, 128009],
)
thread = Thread(target=model.generate, kwargs=generate_kwargs)
thread.start()
buffer = ""
for new_text in streamer:
buffer += new_text
yield buffer
chatbot = gr.Chatbot(height=500)
with gr.Blocks() as demo:
gr.HTML(TITLE)
gr.HTML(DESCRIPTION)
gr.ChatInterface(
fn=stream_chat,
chatbot=chatbot,
fill_height=True,
theme="soft",
retry_btn=None,
undo_btn="Delete Previous",
clear_btn="Clear",
additional_inputs_accordion=gr.Accordion(label="⚙️ Parameters", open=False, render=False),
additional_inputs=[
gr.Slider(
minimum=0,
maximum=1,
step=0.1,
value=0.8,
label="Temperature",
render=False,
),
gr.Slider(
minimum=128,
maximum=4096,
step=1,
value=1024,
label="Max new tokens",
render=False,
),
gr.Slider(
minimum=0.0,
maximum=1.0,
step=0.1,
value=0.8,
label="top_p",
render=False,
),
gr.Slider(
minimum=1,
maximum=20,
step=1,
value=20,
label="top_k",
render=False,
),
gr.Slider(
minimum=0.0,
maximum=2.0,
step=0.1,
value=1.0,
label="Repetition penalty",
render=False,
),
],
examples=[
["Help me study vocabulary: write a sentence for me to fill in the blank, and I'll try to pick the correct option."],
["What are 5 creative things I could do with my kids' art? I don't want to throw them away, but it's also so much clutter."],
["Tell me a random fun fact about the Roman Empire."],
["Show me a code snippet of a website's sticky header in CSS and JavaScript."],
],
cache_examples=False,
)
if __name__ == "__main__":
demo.launch()
|