|
|
import os |
|
|
from threading import Thread |
|
|
from typing import Iterator |
|
|
import time |
|
|
import gradio as gr |
|
|
import spaces |
|
|
import torch |
|
|
import json |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer |
|
|
|
|
|
|
|
|
DESCRIPTION = """\ |
|
|
Shakti is a 2.5 billion parameter language model specifically optimized for resource-constrained environments such as edge devices, including smartphones, wearables, and IoT systems. With support for vernacular languages and domain-specific tasks, Shakti excels in industries such as healthcare, finance, and customer service |
|
|
For more details, please check [here](https://arxiv.org/pdf/2410.11331v1). |
|
|
""" |
|
|
|
|
|
MAX_MAX_NEW_TOKENS = 2048 |
|
|
DEFAULT_MAX_NEW_TOKENS = 1024 |
|
|
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "2048")) |
|
|
|
|
|
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu") |
|
|
|
|
|
model_id = "SandLogicTechnologies/Shakti-2.5B-RoPE-Scale" |
|
|
tokenizer = AutoTokenizer.from_pretrained(model_id, token=os.getenv("SHAKTI")) |
|
|
model = AutoModelForCausalLM.from_pretrained( |
|
|
model_id, |
|
|
device_map="auto", |
|
|
torch_dtype=torch.bfloat16, |
|
|
token=os.getenv("SHAKTI") |
|
|
|
|
|
) |
|
|
model.eval() |
|
|
|
|
|
|
|
|
@spaces.GPU(duration=180) |
|
|
def generate( |
|
|
message: str, |
|
|
chat_history: list[tuple[str, str]], |
|
|
max_new_tokens: int = 1024, |
|
|
temperature: float = 0.6, |
|
|
top_p: float = 0.9, |
|
|
top_k: int = 50, |
|
|
repetition_penalty: float = 1.2, |
|
|
) -> Iterator[str]: |
|
|
conversation = [json.loads(os.getenv("PROMPT"))] |
|
|
|
|
|
for user, assistant in chat_history: |
|
|
conversation.extend( |
|
|
[ |
|
|
|
|
|
{"role": "user", "content": user}, |
|
|
{"role": "assistant", "content": assistant}, |
|
|
] |
|
|
) |
|
|
print("Content: ", user) |
|
|
conversation.append({"role": "user", "content": message}) |
|
|
|
|
|
input_ids = tokenizer.apply_chat_template(conversation, add_generation_prompt=True, return_tensors="pt") |
|
|
if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH: |
|
|
input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:] |
|
|
gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.") |
|
|
input_ids = input_ids.to(model.device) |
|
|
|
|
|
streamer = TextIteratorStreamer(tokenizer, timeout=50.0, skip_prompt=True, skip_special_tokens=True) |
|
|
generate_kwargs = dict( |
|
|
{"input_ids": input_ids}, |
|
|
streamer=streamer, |
|
|
max_new_tokens=max_new_tokens, |
|
|
do_sample=True, |
|
|
top_p=top_p, |
|
|
top_k=top_k, |
|
|
temperature=temperature, |
|
|
num_beams=1, |
|
|
repetition_penalty=repetition_penalty, |
|
|
) |
|
|
t = Thread(target=model.generate, kwargs=generate_kwargs) |
|
|
t.start() |
|
|
|
|
|
outputs = [] |
|
|
for text in streamer: |
|
|
outputs.append(text) |
|
|
yield "".join(outputs) |
|
|
|
|
|
|
|
|
chat_interface = gr.ChatInterface( |
|
|
fn=generate, |
|
|
additional_inputs=[ |
|
|
gr.Slider( |
|
|
label="Max new tokens", |
|
|
minimum=1, |
|
|
maximum=MAX_MAX_NEW_TOKENS, |
|
|
step=1, |
|
|
value=DEFAULT_MAX_NEW_TOKENS, |
|
|
), |
|
|
gr.Slider( |
|
|
label="Temperature", |
|
|
minimum=0.1, |
|
|
maximum=4.0, |
|
|
step=0.1, |
|
|
value=0.6, |
|
|
), |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
], |
|
|
stop_btn=None, |
|
|
examples=[ |
|
|
["Tell me a story"], ["write a short poem which is hard to sing"], ['मुझे भारतीय इतिहास के बारे में बताएं'] |
|
|
], |
|
|
cache_examples=False, |
|
|
) |
|
|
|
|
|
with gr.Blocks(css="style.css", fill_height=True) as demo: |
|
|
gr.Markdown(DESCRIPTION) |
|
|
gr.DuplicateButton(value="Duplicate Space for private use", elem_id="duplicate-button") |
|
|
chat_interface.render() |
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.queue(max_size=20).launch() |
