Spaces:
Sleeping
Sleeping
File size: 3,477 Bytes
afec331 |
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 |
import subprocess
from threading import Thread
import torch
import spaces
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig, TextIteratorStreamer
#subprocess.run('pip install flash-attn --no-build-isolation', env={'FLASH_ATTENTION_SKIP_CUDA_BUILD': "TRUE"}, shell=True)
MODEL_ID = "deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B"
CHAT_TEMPLATE = "َAuto"
MODEL_NAME = MODEL_ID.split("/")[-1]
CONTEXT_LENGTH = 16000
latex_delimiters_set = [{
"left": "\\(",
"right": "\\)",
"display": False
}, {
"left": "\\begin{equation}",
"right": "\\end{equation}",
"display": True
}, {
"left": "\\begin{align}",
"right": "\\end{align}",
"display": True
}, {
"left": "\\begin{alignat}",
"right": "\\end{alignat}",
"display": True
}, {
"left": "\\begin{gather}",
"right": "\\end{gather}",
"display": True
}, {
"left": "\\begin{CD}",
"right": "\\end{CD}",
"display": True
}, {
"left": "\\[",
"right": "\\]",
"display": True
}]
def predict(message, history, system_prompt, temperature, max_new_tokens, top_k, repetition_penalty, top_p):
# Format history with a given chat template
stop_tokens = ["<|endoftext|>", "<|im_end|>","|im_end|"]
instruction = '<|im_start|>system\n' + system_prompt + '\n<|im_end|>\n'
for user, assistant in history:
instruction += f'<|im_start|>user\n{user}\n<|im_end|>\n<|im_start|>assistant\n{assistant}\n<|im_end|>\n'
instruction += f'<|im_start|>user\n{message}\n<|im_end|>\n<|im_start|>assistant\n'
print(instruction)
streamer = TextIteratorStreamer(tokenizer, skip_prompt=True, skip_special_tokens=False)
enc = tokenizer(instruction, return_tensors="pt", padding=True, truncation=True)
input_ids, attention_mask = enc.input_ids, enc.attention_mask
if input_ids.shape[1] > CONTEXT_LENGTH:
input_ids = input_ids[:, -CONTEXT_LENGTH:]
attention_mask = attention_mask[:, -CONTEXT_LENGTH:]
generate_kwargs = dict(
input_ids=input_ids.to(device),
attention_mask=attention_mask.to(device),
streamer=streamer,
do_sample=True,
temperature=temperature,
max_new_tokens=max_new_tokens,
top_k=top_k,
repetition_penalty=repetition_penalty,
top_p=top_p
)
t = Thread(target=model.generate, kwargs=generate_kwargs)
t.start()
outputs = []
for new_token in streamer:
outputs.append(new_token)
if new_token in stop_tokens:
break
yield "".join(outputs)
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
model = AutoModelForCausalLM.from_pretrained(
MODEL_ID,
device_map="auto",
)
# Create Gradio interface
gr.ChatInterface(
predict,
additional_inputs_accordion=gr.Accordion(label="Parameters", open=False),
additional_inputs=[
gr.Textbox("You are a useful assistant. first recognize user request and then reply carfuly and thinking", label="System prompt"),
gr.Slider(0, 1, 0.6, label="Temperature"),
gr.Slider(0, 32000, 10000, label="Max new tokens"),
gr.Slider(1, 80, 40, label="Top K sampling"),
gr.Slider(0, 2, 1.1, label="Repetition penalty"),
gr.Slider(0, 1, 0.95, label="Top P sampling"),
],
).queue().launch()
|