Spaces:
Sleeping
Sleeping
| ''' | |
| import gradio as gr | |
| from transformers import AutoTokenizer, AutoModelForTokenClassification | |
| import torch | |
| #Load model and tokenizer | |
| #model_name = "Shenkezs/nbme-deberta-v3-finetuned" | |
| #tokenizer_name = "microsoft/deberta-v3-base" | |
| model_name = "smeoni/nbme-Bio_ClinicalBERT" | |
| tokenizer = AutoTokenizer.from_pretrained(model_name) | |
| model = AutoModelForTokenClassification.from_pretrained(model_name) | |
| #tokenizer = AutoTokenizer.from_pretrained(model_name) | |
| #model = AutoModelForTokenClassification.from_pretrained(model_name) | |
| # Define prediction function | |
| def predict(text): | |
| inputs = tokenizer(text, return_tensors="pt", truncation=True) | |
| with torch.no_grad(): | |
| outputs = model(**inputs).logits | |
| predictions = torch.argmax(outputs, dim=-1) | |
| tokens = tokenizer.convert_ids_to_tokens(inputs["input_ids"][0]) | |
| labels = predictions[0].tolist() | |
| return {t: l for t, l in zip(tokens, labels)} | |
| # Create Gradio Interface | |
| demo = gr.Interface( | |
| fn=predict, | |
| inputs=gr.Textbox(lines=3, placeholder="Enter patient note or text...."), | |
| outputs="json", | |
| title="NBME DeBERTa Fine-tuned Model", | |
| description="Predicts token-level classifications from the fine-tuned NBME model." | |
| ) | |
| if __name__ == "__main__": | |
| demo.launch() | |
| ''' | |
| import gradio as gr | |
| from transformers import pipeline | |
| # Load model pipeline | |
| pipe = pipeline("text-generation", model="Orion-zhen/Qwen2.5-7B-Instruct-Uncensored") | |
| # Chat function | |
| def chat_fn(message, history): | |
| response = pipe(message, max_new_tokens=250, do_sample=True, temperature=0.7)[0]['generated_text'] | |
| return response.strip() | |
| # Custom theme and layout | |
| with gr.Blocks( | |
| title="NBME DeBERTa Finetuned Assistant", | |
| theme=gr.themes.Soft(primary_hue="blue"), | |
| css=""" | |
| .gradio-container { | |
| font-family: 'Inter', sans-serif; | |
| background-color: white; | |
| color: #1a1a1a; | |
| } | |
| .gr-button { | |
| border-radius: 8px !important; | |
| } | |
| #chatbot { | |
| background-color: white; | |
| border-radius: 12px; | |
| box-shadow: 0 2px 8px rgba(0,0,0,0.05); | |
| } | |
| .svelte-drum-input textarea, .svelte-drum-input input { | |
| background-color: white !important; | |
| color: black !important; | |
| } | |
| .svelte-drum-examples button { | |
| background-color: white !important; | |
| color: black !important; | |
| border: 1px solid #ccc !important; | |
| } | |
| .svelte-drum-examples button:hover { | |
| background-color: #f5f5f5 !important; | |
| } | |
| footer {visibility: hidden;} | |
| """ | |
| ) as demo: | |
| with gr.Row(): | |
| with gr.Column(scale=1, min_width=250): | |
| with gr.Group(): | |
| gr.Markdown(""" | |
| ### 🧠 NBME DeBERTa Finetuned Chatbot | |
| A research-focused conversational assistant powered by a fine-tuned transformer model. | |
| Use this app to explore clinical, academic, or conceptual discussions. | |
| """) | |
| gr.Markdown("**Sign in** to use this API with your Hugging Face account.") | |
| login_button = gr.LoginButton("🔐 Sign in with Hugging Face") | |
| gr.Markdown("---") | |
| gr.Markdown(""" | |
| #### About | |
| This chatbot is based on a fine-tuned DeBERTa model trained on the NBME dataset. | |
| It supports text-based reasoning, knowledge queries, and conceptual explanations. | |
| """) | |
| gr.Markdown("[📄 View Model Card](https://huggingface.co/Shenkezs/nbme-deberta-v3-finetuned)") | |
| with gr.Column(scale=3): | |
| gr.ChatInterface( | |
| fn=chat_fn, | |
| title="Chat Interface", | |
| description="Engage in research-level discussions or ask general knowledge questions.", | |
| chatbot=gr.Chatbot(label="Academic Research Assistant"), | |
| examples=[ | |
| ["Explain gravity to a 5-year-old."], | |
| ["What were the main causes of World War I?"], | |
| ["Tell me a joke about calculus."], | |
| ["What is the capital of Pakistan?"] | |
| ] | |
| ) | |
| demo.launch(share=True) |