app.py

import spaces
import gradio as gr
from transformers import AutoModelForCausalLM, AutoTokenizer, AutoModelForSequenceClassification, BitsAndBytesConfig
import torch
from gradio_rich_textbox import RichTextbox

title = """# Welcome to 🌟Tonic's🧬📏💪🏻Genstruct 7B !
🧬📏💪🏻[Genstruct 7B](https://huggingface.co/NousResearch/Genstruct-7B) is an instruction-generation model, designed to create valid instructions given a raw text corpus. This enables the creation of new, partially synthetic instruction finetuning datasets from any raw-text corpus. You can build with this endpoint using🧬📏💪🏻[Genstruct 7B](https://huggingface.co/NousResearch/Genstruct-7B) available here : [NousResearch/Genstruct-7B](https://huggingface.co/NousResearch/Genstruct-7B). You can also use ✨StarCoder by cloning this space. Simply click here: <a style="display:inline-block" href="https://huggingface.co/spaces/Tonic/starcoder2?duplicate=true"><img src="https://img.shields.io/badge/-Duplicate%20Space-blue?labelColor=white&style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAABAAAAAQCAYAAAAf8/9hAAAAAXNSR0IArs4c6QAAAP5JREFUOE+lk7FqAkEURY+ltunEgFXS2sZGIbXfEPdLlnxJyDdYB62sbbUKpLbVNhyYFzbrrA74YJlh9r079973psed0cvUD4A+4HoCjsA85X0Dfn/RBLBgBDxnQPfAEJgBY+A9gALA4tcbamSzS4xq4FOQAJgCDwV2CPKV8tZAJcAjMMkUe1vX+U+SMhfAJEHasQIWmXNN3abzDwHUrgcRGmYcgKe0bxrblHEB4E/pndMazNpSZGcsZdBlYJcEL9Afo75molJyM2FxmPgmgPqlWNLGfwZGG6UiyEvLzHYDmoPkDDiNm9JR9uboiONcBXrpY1qmgs21x1QwyZcpvxt9NS09PlsPAAAAAElFTkSuQmCC&logoWidth=14" alt="Duplicate Space"></a></h3> 
Join us : 🌟TeamTonic🌟 is always making cool demos! Join our active builder's 🛠️community 👻 [![Join us on Discord](https://img.shields.io/discord/1109943800132010065?label=Discord&logo=discord&style=flat-square)](https://discord.gg/GWpVpekp) On 🤗Huggingface:[MultiTransformer](https://huggingface.co/MultiTransformer) Math 🔍 [introspector](https://huggingface.co/introspector) On 🌐Github: [Tonic-AI](https://github.com/tonic-ai) & contribute to🌟 [SciTonic](https://github.com/Tonic-AI/multitonic)🤗Big thanks to Yuvi Sharma and all the folks at huggingface for the community grant 🤗
"""

examplecofee = """A cortado is a Spanish beverage consisting of espresso mixed with a roughly equal amount of warm milk to reduce the acidity,[1][2] although the exact ratios have considerable regional variation.[3] The milk in a cortado is steamed, but not frothy and "texturized" as in many Italian coffee drinks.[4] The cortado is commonly served all over Spain.[5] The word cortado is the past participle of the Spanish verb cortar (to cut), in the sense of "dilute", and can refer variously to either coffee or espresso drinks throughout Spanish and Portuguese speaking countries."""

model_path = "NousResearch/Genstruct-7B"

tokenizer = AutoTokenizer.from_pretrained(model_path)
quantization_config = BitsAndBytesConfig(load_in_8bit=True)
model = AutoModelForCausalLM.from_pretrained(model_path, device_map='cuda', quantization_config=quantization_config)
rm_tokenizer = AutoTokenizer.from_pretrained('OpenAssistant/reward-model-deberta-v3-large-v2')
rm_model = AutoModelForSequenceClassification.from_pretrained('OpenAssistant/reward-model-deberta-v3-large-v2', device_map='cuda', torch_dtype=torch.bfloat16)


@spaces.GPU
def generate_text(usertitle, content, temperature, max_length, N=3):
    # msg ={
    # 'title': usertitle,
    # 'content': content
    # }
    input_text = f"[[[title:]]] {usertitle}\n[[[content:]]]{content}\n\n"
    inputs = tokenizer(input_text, return_tensors='pt').to('cuda')
    attention_mask = torch.ones(inputs['input_ids'].shape, dtype=torch.long, device='cuda')
    generated_sequences = model.generate(inputs['input_ids'], attention_mask=attention_mask, temperature=temperature, max_length=max_length, pad_token_id=tokenizer.eos_token_id, num_return_sequences=N, do_sample=True)
    decoded_sequences = [tokenizer.decode(g, skip_special_tokens=True) for g in generated_sequences]
    def score(sequence):
        inputs = rm_tokenizer(sequence, return_tensors='pt', padding=True, truncation=True, max_length=512).to('cuda')
        inputs = {k: v.to('cuda') for k, v in inputs.items()} 
#       print(rm_model.device)
#       print(inputs['input_ids'].device) 
        with torch.no_grad():
            outputs = rm_model(**inputs)
            logits = outputs.logits
            print("Logits shape:", logits.shape)  
            print("Logits contents:", logits) 
            return logits[0]

    best_sequence = max(decoded_sequences, key=score)

    return best_sequence

def gradio_app():
    with gr.Blocks() as demo:
        gr.Markdown(title)
        usertitle = gr.Textbox(label="Title", value="Cortado", lines=1)
        content = gr.Textbox(label="WordPhrases", value=examplecofee, lines=5)
        with gr.Row():
            temperature = gr.Slider(minimum=0.1, maximum=1.0, step=0.1, value=0.5, label="Temperature")
            max_length = gr.Slider(minimum=250, maximum=1024, step=10, value=450, label="Generate Length")
        generate_btn = gr.Button("Try 🧬📏💪🏻 Genstruct")
        output = gr.Textbox(label="🧬📏💪🏻Genstruct 7B:")

        generate_btn.click(
            fn=generate_text,
            inputs=[usertitle, content, temperature, max_length],
            outputs=output
        )

    demo.launch()

if __name__ == "__main__":
    gradio_app()