Update app.py

app.py

@@ -1,113 +1,76 @@
-import json
-import gradio as gr
-from openai import OpenAI
 from langchain_community.vectorstores import FAISS
 from langchain_community.embeddings import FakeEmbeddings
-from langchain.docstore.document import Document
 from langchain.text_splitter import CharacterTextSplitter
+from langchain_community.document_loaders import TextLoader
+from openai import OpenAI
+
+import os
+import gradio as gr
+import json
+
+# ========== Load Documents and Create Vector DB ==========
+loader = TextLoader("mindmate.txt")
+documents = loader.load()
 
-# --- DeepSeek R1 API Setup via OpenRouter ---
+text_splitter = CharacterTextSplitter(chunk_size=1000, chunk_overlap=100)
+split_docs = text_splitter.split_documents(documents)
+
+# Fix for FakeEmbeddings requiring size
+embedding = FakeEmbeddings(size=1536)
+vector_db = FAISS.from_documents(split_docs, embedding)
+
+# ========== OpenRouter DeepSeek R1 Setup ==========
 client = OpenAI(
     base_url="https://openrouter.ai/api/v1",
     api_key="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",
 )
 
-# --- Dummy Embeddings and RAG Docs ---
-documents = [
-    Document(page_content="Take a deep breath. You are doing your best."),
-    Document(page_content="It's okay to not be okay. Give yourself grace."),
-    Document(page_content="You are not alone. Many have walked this path and found light again."),
-    Document(page_content="Small steps lead to big changes. Just begin.")
-]
-splitter = CharacterTextSplitter(chunk_size=100, chunk_overlap=0)
-split_docs = splitter.split_documents(documents)
-vector_db = FAISS.from_documents(split_docs, FakeEmbeddings())
-
-# --- Emotion Detection using DeepSeek R1 ---
-def detect_emotion(user_input):
-    prompt = f"""
-    Analyze the message and return a JSON with:
-    {{
-      "emotion": "<Emotion>",
-      "tone": "<Suggested_Tone>",
-      "affirmation": "<One_Line_Affirmation>",
-      "response": "<Comforting response>"
-    }}
-    Message: "{user_input}"
-    """
+# ========== Gradio App ==========
+def mindmate(name, feeling):
+    if not name or not feeling:
+        return "Please enter your name and what's troubling you."
 
-    completion = client.chat.completions.create(
-        model="deepseek/deepseek-r1:free",
-        messages=[{"role": "user", "content": prompt}]
-    )
-    return json.loads(completion.choices[0].message.content)
+    context_docs = vector_db.similarity_search(feeling, k=2)
+    context_text = "\n".join([doc.page_content for doc in context_docs])
 
-# --- Extra Tools ---
-def get_affirmation():
-    completion = client.chat.completions.create(
-        model="deepseek/deepseek-r1:free",
-        messages=[{"role": "user", "content": "Give a one-line calming affirmation."}]
-    )
-    return completion.choices[0].message.content
+    prompt = f"""You are a compassionate and empathetic mental health support assistant named MindMate.
+The user's name is {name}. They said: "{feeling}"
 
-def get_journaling_prompt():
-    completion = client.chat.completions.create(
-        model="deepseek/deepseek-r1:free",
-        messages=[{"role": "user", "content": "Give a journaling prompt for mental clarity."}]
-    )
-    return completion.choices[0].message.content
+Based on the context and user's input, respond with a warm, personalized, emotionally intelligent response.
+Use the following context to help, but DO NOT mention it's from a document:
+
+Context:
+{context_text}
+"""
 
-def get_calming_technique():
     completion = client.chat.completions.create(
         model="deepseek/deepseek-r1:free",
-        messages=[{"role": "user", "content": "Suggest a calming breathing or grounding technique."}]
+        messages=[{"role": "user", "content": prompt}],
+        extra_headers={
+            "HTTP-Referer": "https://huggingface.co/spaces/dhruvilhere/mindmate-chatbot",
+            "X-Title": "MindMate Chatbot",
+        },
     )
+
     return completion.choices[0].message.content
 
-# --- Chatbot Main Logic ---
-def chatbot_interface(name, issue):
-    emotion_data = detect_emotion(issue)
-    rag_support = "\n".join([d.page_content for d in vector_db.similarity_search(issue, k=2)])
-
-    system_prompt = f"""
-    You're MindMate, a kind mental health assistant.
-    Talk directly to {name}.
-    Emotion: {emotion_data['emotion']}
-    Tone: {emotion_data['tone']}
-    Affirmation: {emotion_data['affirmation']}
-    Message: {issue}
-    Based on this and the below RAG support:
-    {rag_support}
-    Write a warm, comforting response.
-    """
-
-    final_response = client.chat.completions.create(
-        model="deepseek/deepseek-r1:free",
-        messages=[{"role": "user", "content": system_prompt}]
-    )
 
-    return {
-        "Emotion": emotion_data['emotion'],
-        "Affirmation": emotion_data['affirmation'],
-        "Companion Response": final_response.choices[0].message.content,
-        "Journaling Prompt": get_journaling_prompt(),
-        "Calming Tip": get_calming_technique()
-    }
-
-# --- Gradio UI ---
-inputs = [
-    gr.Textbox(label="Your Name", placeholder="e.g. Dhruvil"),
-    gr.Textbox(label="What's troubling you today?", lines=4)
-]
-
-outputs = gr.JSON(label="🧠 MindMate's Support")
-
-demo = gr.Interface(
-    fn=chatbot_interface,
-    inputs=inputs,
-    outputs=outputs,
-    title="🧠 MindMate: Your Mental Health Companion",
-    description="Just share your name and how you're feeling. MindMate will support you emotionally with warmth, care, and science. Powered by DeepSeek R1 via OpenRouter 💖"
-)
+# ========== Gradio UI ==========
+with gr.Blocks(theme=gr.themes.Soft(), ssr=True) as demo:
+    gr.Markdown("🧠 **MindMate: Your Mental Health Companion**")
+    gr.Markdown("Just share your name and how you're feeling. MindMate will support you emotionally with warmth, care, and science. No key needed ❤️")
+
+    with gr.Row():
+        with gr.Column():
+            name = gr.Textbox(label="Your Name")
+            feeling = gr.Textbox(label="What's troubling you today?", lines=4)
+            submit_btn = gr.Button("Submit", scale=1)
+            clear_btn = gr.Button("Clear")
+
+        with gr.Column():
+            output = gr.JSON(label="🧘 MindMate's Support")
+
+    submit_btn.click(fn=mindmate, inputs=[name, feeling], outputs=output)
+    clear_btn.click(fn=lambda: ("", "", {}), inputs=[], outputs=[name, feeling, output])
 
-demo.launch()
+demo.launch()