Upload 6 files

README.md

@@ -1,13 +1,42 @@
----
-title: FeelMate
-emoji: 📈
-colorFrom: indigo
-colorTo: purple
-sdk: gradio
-sdk_version: 5.25.2
-app_file: app.py
-pinned: false
-license: mit
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# FeelMate 🤗
+
+FeelMate is a compassionate chatbot designed to provide emotional support and a safe space for users to express their feelings without judgment.
+
+## Features
+
+- 🤝 Empathetic and understanding responses
+- 🔒 Private and secure conversations
+- 💬 Natural and supportive dialogue
+- 🧠 Memory of previous conversations
+- 🎨 Clean and intuitive interface
+
+## Setup
+
+1. Clone this repository
+2. Install the required dependencies:
+   ```bash
+   pip install -r requirements.txt
+   ```
+3. Create a `.env` file in the root directory and add your Hugging Face API token:
+   ```
+   HUGGINGFACEHUB_API_TOKEN=your_api_token_here
+   ```
+4. Run the application:
+   ```bash
+   python app.py
+   ```
+
+## How to Use
+
+1. Launch the application
+2. Start typing your feelings or concerns in the chat interface
+3. FeelMate will respond with understanding and support
+4. Continue the conversation as needed
+
+## Note
+
+While FeelMate is designed to provide emotional support, it is not a replacement for professional mental health services. If you're experiencing severe emotional distress, please consider reaching out to a mental health professional.
+
+## License
+
+This project is licensed under the MIT License - see the LICENSE file for details. 

app.py

@@ -0,0 +1,33 @@
+import gradio as gr
+from chain import get_feelmate_chain
+
+# Load the LangChain-powered FeelMate
+feelmate_chain = get_feelmate_chain()
+
+def chat_with_feelmate(user_message, history):
+    if not user_message.strip():
+        return "Please share something with me 💬", history
+    
+    # Add user input to memory and get the response
+    response = feelmate_chain.predict(input=user_message)
+    
+    # Add the current conversation to the history
+    history.append((user_message, response))
+    
+    # Limit the number of history items if it's too long
+    history = history[-10:]  # Keep last 10 exchanges
+    
+    return response.strip(), history
+
+# Gradio interface with memory to display history
+iface = gr.Interface(
+    fn=chat_with_feelmate,
+    inputs=[gr.Textbox(lines=3, placeholder="Tell me how you're feeling today..."), gr.State()],
+    outputs=[gr.Textbox(), gr.State()],
+    title="🫂 FeelMate - Your Emotional Support Buddy",
+    description="No judgment. Just kindness, understanding, and a little light. 💜",
+    live=True
+)
+
+if __name__ == "__main__":
+    iface.launch()

chain.py

@@ -0,0 +1,18 @@
+from langchain.llms import HuggingFacePipeline
+from langchain.chains import ConversationChain
+from langchain.memory import ConversationBufferMemory
+from model_loader import load_openchat
+
+def get_feelmate_chain():
+    pipe = load_openchat()
+    llm = HuggingFacePipeline(pipeline=pipe)
+
+    memory = ConversationBufferMemory()  # stores chat history in memory
+
+    chain = ConversationChain(
+        llm=llm,
+        memory=memory,
+        verbose=False  # You can set to True for debugging
+    )
+
+    return chain

model_loader.py

@@ -0,0 +1,30 @@
+
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from langchain_community.llms import HuggingFacePipeline
+from langchain.chains import ConversationChain
+from langchain.memory import ConversationBufferMemory
+from transformers import pipeline
+
+def load_openchat():
+    # Load the Llama model and tokenizer
+    model_name = "meta-llama/Llama-2-7b-chat-hf"  # Use your Llama model path or Hugging Face model ID
+    model = AutoModelForCausalLM.from_pretrained(model_name)
+    tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+    # Create a pipeline for text generation
+    pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
+
+    # Use LangChain to create a conversation chain with memory
+    llm = HuggingFacePipeline(pipeline=pipe)
+    
+    # Set up memory to store the conversation history
+    memory = ConversationBufferMemory()
+    
+    # Set up the conversation chain
+    conversation_chain = ConversationChain(
+        llm=llm,
+        memory=memory,
+        verbose=True
+    )
+
+    return conversation_chain

prompt_template.py

@@ -0,0 +1,12 @@
+from langchain.prompts import PromptTemplate
+
+feelmate_prompt = PromptTemplate(
+    input_variables=["user_input"],
+    template="""
+You are FeelMate, a kind and empathetic AI friend. You never judge, always listen, and deeply care about the user's emotions.
+
+Respond supportively, reflectively, and in a friendly tone. Your goal is to make the user feel heard, understood, and a little better.
+
+User: {user_input}
+FeelMate:"""
+)

requirements.txt

@@ -0,0 +1,6 @@
+langchain
+transformers
+torch
+accelerate
+gradio
+dotenv