Update app.py

app.py

@@ -1,70 +1,70 @@
-# import gradio as gr
-# from huggingface_hub import InferenceClient
-# import os
+import gradio as gr
+from huggingface_hub import InferenceClient
+import os
 
-# """
-# For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-# """
+"""
+For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
+"""
 
-# token = os.getenv("HF_TOKEN")  
-# client = InferenceClient("BioMistral/BioMistral-7B", token=token)
+token = os.getenv("HF_TOKEN")  
+client = InferenceClient("emilyalsentzer/Bio_ClinicalBERT", token=token)
 
 
-# def respond(
-#     message,
-#     history: list[tuple[str, str]],
-#     system_message,
-#     max_tokens,
-#     temperature,
-#     top_p,
-# ):
-#     messages = [{"role": "system", "content": system_message}]
+def respond(
+    message,
+    history: list[tuple[str, str]],
+    system_message,
+    max_tokens,
+    temperature,
+    top_p,
+):
+    messages = [{"role": "system", "content": system_message}]
 
-#     for val in history:
-#         if val[0]:
-#             messages.append({"role": "user", "content": val[0]})
-#         if val[1]:
-#             messages.append({"role": "assistant", "content": val[1]})
+    for val in history:
+        if val[0]:
+            messages.append({"role": "user", "content": val[0]})
+        if val[1]:
+            messages.append({"role": "assistant", "content": val[1]})
 
-#     messages.append({"role": "user", "content": message})
+    messages.append({"role": "user", "content": message})
 
-#     response = ""
+    response = ""
 
-#     for message in client.chat_completion(
-#         messages,
-#         max_tokens=max_tokens,
-#         stream=True,
-#         temperature=temperature,
-#         top_p=top_p,
-#     ):
-#         token = message.choices[0].delta.content
+    for message in client.chat_completion(
+        messages,
+        max_tokens=max_tokens,
+        stream=True,
+        temperature=temperature,
+        top_p=top_p,
+    ):
+        token = message.choices[0].delta.content
 
-#         response += token
-#         yield response
+        response += token
+        yield response
 
 
-# """
-# For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-# """
-# demo = gr.ChatInterface(
-#     respond,
-#     additional_inputs=[
-#         gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-#         gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-#         gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-#         gr.Slider(
-#             minimum=0.1,
-#             maximum=1.0,
-#             value=0.95,
-#             step=0.05,
-#             label="Top-p (nucleus sampling)",
-#         ),
-#     ],
-# )
+"""
+For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
+"""
+demo = gr.ChatInterface(
+    respond,
+    additional_inputs=[
+        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
+        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
+        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.95,
+            step=0.05,
+            label="Top-p (nucleus sampling)",
+        ),
+    ],
+)
 
 
-# if __name__ == "__main__":
-#     demo.launch()
+if __name__ == "__main__":
+    demo.launch()
 
 
 # import gradio as gr
@@ -170,70 +170,70 @@
 #     iface.launch()
 
 
-import gradio as gr
-from langdetect import detect
-from transformers import AutoTokenizer, AutoModelForCausalLM, TextGenerationPipeline, GenerationConfig
-import torch
+# import gradio as gr
+# from langdetect import detect
+# from transformers import AutoTokenizer, AutoModelForCausalLM, TextGenerationPipeline, GenerationConfig
+# import torch
 
-# Load model and tokenizer
-model_name = "FreedomIntelligence/Apollo-7B"
+# # Load model and tokenizer
+# model_name = "FreedomIntelligence/Apollo-7B"
 
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name)
+# tokenizer = AutoTokenizer.from_pretrained(model_name)
+# model = AutoModelForCausalLM.from_pretrained(model_name)
 
-tokenizer.pad_token = tokenizer.eos_token
+# tokenizer.pad_token = tokenizer.eos_token
 
-generation_config = GenerationConfig(
-    max_new_tokens=150,
-    temperature=0.2,
-    top_k=20,
-    do_sample=True,
-    top_p=0.7,
-    repetition_penalty=1.3,
-)
+# generation_config = GenerationConfig(
+#     max_new_tokens=150,
+#     temperature=0.2,
+#     top_k=20,
+#     do_sample=True,
+#     top_p=0.7,
+#     repetition_penalty=1.3,
+# )
 
-# Create generation pipeline
-pipe = TextGenerationPipeline(
-    model=model,
-    tokenizer=tokenizer,
-    device=model.device.index if torch.cuda.is_available() else "cpu"
-)
+# # Create generation pipeline
+# pipe = TextGenerationPipeline(
+#     model=model,
+#     tokenizer=tokenizer,
+#     device=model.device.index if torch.cuda.is_available() else "cpu"
+# )
 
-# Prompt formatter based on language
-def generate_prompt(message, history):
-    lang = detect(message)
-    if lang == "ar":
-        return f"""أجب على السؤال الطبي التالي بلغة عربية فصحى، بإجابة دقيقة ومفصلة. إذا لم تجد معلومات كافية في السياق، استخدم معرفتك الطبية السابقة.
- وتأكد من ان:
-- عدم تكرار أي نقطة أو عبارة أو كلمة
-- وضوح وسلاسة كل نقطة
-- تجنب الحشو والعبارات الزائدة
-
-السؤال: {message}
-الإجابة:"""
-    else:
-        return f"""Answer the following medical question in clear English with a detailed, non-redundant response. Do not repeat ideas or restate the question. If information is missing, rely on your prior medical knowledge:
-Question: {message}
-Answer:"""
-
-# Chat function
-def chat_fn(message, history):
-    prompt = generate_prompt(message, history)
-    response = pipe(prompt,
-                    max_new_tokens=512,
-                    temperature=0.7,
-                    do_sample = True,
-                    top_p=0.9)[0]['generated_text']
-    answer = response.split("Answer:")[-1].strip() if "Answer:" in response else response.split("الإجابة:")[-1].strip()
-    return answer
-
-# Gradio ChatInterface
-demo = gr.ChatInterface(
-    fn=chat_fn,
-    title="🩺 Apollo-7B Medical Chatbot",
-    description="Multilingual (Arabic & English) medical Q&A chatbot powered by Apollo-7B. No RAG, just fast model inference.",
-    theme=gr.themes.Soft()
-)
+# # Prompt formatter based on language
+# def generate_prompt(message, history):
+#     lang = detect(message)
+#     if lang == "ar":
+#         return f"""أجب على السؤال الطبي التالي بلغة عربية فصحى، بإجابة دقيقة ومفصلة. إذا لم تجد معلومات كافية في السياق، استخدم معرفتك الطبية السابقة.
+#  وتأكد من ان:
+# - عدم تكرار أي نقطة أو عبارة أو كلمة
+# - وضوح وسلاسة كل نقطة
+# - تجنب الحشو والعبارات الزائدة
 
-if __name__ == "__main__":
-    demo.launch(share=True)
+# السؤال: {message}
+# الإجابة:"""
+#     else:
+#         return f"""Answer the following medical question in clear English with a detailed, non-redundant response. Do not repeat ideas or restate the question. If information is missing, rely on your prior medical knowledge:
+# Question: {message}
+# Answer:"""
+
+# # Chat function
+# def chat_fn(message, history):
+#     prompt = generate_prompt(message, history)
+#     response = pipe(prompt,
+#                     max_new_tokens=512,
+#                     temperature=0.7,
+#                     do_sample = True,
+#                     top_p=0.9)[0]['generated_text']
+#     answer = response.split("Answer:")[-1].strip() if "Answer:" in response else response.split("الإجابة:")[-1].strip()
+#     return answer
+
+# # Gradio ChatInterface
+# demo = gr.ChatInterface(
+#     fn=chat_fn,
+#     title="🩺 Apollo-7B Medical Chatbot",
+#     description="Multilingual (Arabic & English) medical Q&A chatbot powered by Apollo-7B. No RAG, just fast model inference.",
+#     theme=gr.themes.Soft()
+# )
+
+# if __name__ == "__main__":
+#     demo.launch(share=True)