Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+faiss_index/index.faiss filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,2 @@
+.env
+__pycache__/

README.md

@@ -1,16 +1,17 @@
 ---
 title: Deater Chat
-emoji: 💬
+emoji: 🧬
 colorFrom: yellow
 colorTo: purple
 sdk: gradio
 sdk_version: 5.42.0
 app_file: app.py
 pinned: false
-hf_oauth: true
-hf_oauth_scopes:
-- inference-api
 license: mit
 ---
 
-An example chatbot using [Gradio](https://gradio.app), [`huggingface_hub`](https://huggingface.co/docs/huggingface_hub/v0.22.2/en/index), and the [Hugging Face Inference API](https://huggingface.co/docs/api-inference/index).
+# HSAN1 Research Assistant
+
+A RAG-powered chatbot that helps patients and families understand HSAN1 (Hereditary Sensory and Autonomic Neuropathy Type 1) using 246 research documents.
+
+Built with [Gradio](https://gradio.app), [LangChain](https://langchain.com), and Google Gemini.

app.py

@@ -1,70 +1,135 @@
+"""
+HSAN1 Research Assistant - Gradio Chat Interface with RAG
+"""
+import os
 import gradio as gr
-from huggingface_hub import InferenceClient
+from dotenv import load_dotenv
+from langchain_google_genai import ChatGoogleGenerativeAI
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_community.vectorstores import FAISS
+from langchain_core.messages import HumanMessage, SystemMessage
+
+# Load environment variables (for local development)
+load_dotenv()
+
+# Configuration
+INDEX_PATH = "./faiss_index"
+SYSTEM_PROMPT = """You are a compassionate medical research assistant helping patients and families understand HSAN1 (Hereditary Sensory and Autonomic Neuropathy Type 1).
+
+You have access to a database of 246 research documents including papers, newsletters, and family histories.
+
+Instructions:
+- Answer questions based ONLY on the provided context
+- If the answer is not in the context, say "I don't see that information in the research documents I have."
+- Use clear, empathetic language and explain medical terms
+- Be accurate but hopeful in tone
+- Keep responses concise but informative"""
+
+# Check for API key
+api_key = os.environ.get("GOOGLE_API_KEY")
+if not api_key:
+    raise ValueError("GOOGLE_API_KEY environment variable not set")
+
+# Load components at startup
+print("Loading embeddings model...")
+embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
+
+print("Loading FAISS index...")
+if not os.path.exists(INDEX_PATH):
+    raise FileNotFoundError(f"FAISS index not found at {INDEX_PATH}. Run build_index.py first.")
+vectorstore = FAISS.load_local(INDEX_PATH, embeddings, allow_dangerous_deserialization=True)
+retriever = vectorstore.as_retriever(search_kwargs={"k": 5})
+
+print("Initializing Gemini...")
+llm = ChatGoogleGenerativeAI(
+    model="gemini-3-flash-preview",
+    temperature=0.3,
+    streaming=True
+)
+
+print("Ready!")
 
 
 def respond(
     message,
     history: list[dict[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-    hf_token: gr.OAuthToken,
 ):
     """
-    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
+    Process a user message and generate a response using RAG.
+    Uses OpenAI-style message format for Gradio ChatInterface.
     """
-    client = InferenceClient(token=hf_token.token, model="openai/gpt-oss-20b")
+    # Retrieve relevant documents
+    docs = retriever.invoke(message)
+    context = "\n\n---\n\n".join([doc.page_content for doc in docs])
 
-    messages = [{"role": "system", "content": system_message}]
+    # Get unique sources
+    sources = list(set([
+        os.path.basename(doc.metadata.get("source", "Unknown"))
+        for doc in docs
+    ]))
 
-    messages.extend(history)
+    # Build the prompt with context
+    augmented_prompt = f"""Context from research documents:
 
-    messages.append({"role": "user", "content": message})
+{context}
 
-    response = ""
+---
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        choices = message.choices
-        token = ""
-        if len(choices) and choices[0].delta.content:
-            token = choices[0].delta.content
+User question: {message}
 
-        response += token
-        yield response
+Please answer based on the context above."""
 
+    # Build messages for the LLM
+    messages = [
+        SystemMessage(content=SYSTEM_PROMPT),
+        HumanMessage(content=augmented_prompt)
+    ]
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
+    # Stream the response
+    response = ""
+    for chunk in llm.stream(messages):
+        if chunk.content:
+            content = chunk.content
+            # Handle different content formats from Gemini models
+            if isinstance(content, list):
+                # Extract text from list of content blocks
+                text_parts = []
+                for item in content:
+                    if isinstance(item, dict) and 'text' in item:
+                        text_parts.append(item['text'])
+                    elif isinstance(item, str):
+                        text_parts.append(item)
+                content = "".join(text_parts)
+            elif isinstance(content, dict) and 'text' in content:
+                content = content['text']
+            response += content
+            yield response
+
+    # Append sources after streaming completes
+    if sources:
+        source_text = f"\n\n---\n*Sources: {', '.join(sources[:3])}*"
+        yield response + source_text
+
+
+# Create Gradio ChatInterface (matching HF template format)
 chatbot = gr.ChatInterface(
     respond,
     type="messages",
-    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)",
-        ),
+    title="🧬 HSAN1 Research Assistant",
+    description="Ask questions about HSAN1 research, treatments, clinical trials, and more. Powered by 246 research documents.",
+    examples=[
+        "What is HSAN1?",
+        "What causes HSAN1?",
+        "Tell me about the L-serine treatment",
+        "What are the symptoms of HSAN1?",
+        "Is there a clinical trial for HSAN1?",
     ],
+    theme="soft",
 )
 
+# Wrap in Blocks for proper rendering
 with gr.Blocks() as demo:
-    with gr.Sidebar():
-        gr.LoginButton()
     chatbot.render()
 
-
 if __name__ == "__main__":
     demo.launch()

faiss_index/index.faiss

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75bcc650fcffe2e8e7c34cf6f3a1f643839bcffaec3286542f187c1df45dd204
+size 2648109

faiss_index/index.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:686886b4f39a31cdd43b6e9e690c4b4bcc35692cb1d1cb5068a7811991f475af
+size 1532857

requirements.txt

@@ -0,0 +1,9 @@
+gradio>=5.0.0
+langchain>=1.2.0
+langchain-community>=0.4.1
+langchain-google-genai>=4.1.2
+langchain-huggingface>=0.1.2
+langchain-text-splitters>=1.1.0
+sentence-transformers>=3.3.0
+faiss-cpu>=1.13.2
+python-dotenv>=1.0.1