Upload 6 files

Dockerfile

@@ -0,0 +1,18 @@
+FROM ghcr.io/astral-sh/uv:python3.13-bookworm-slim
+
+RUN useradd -m -u 1000 user
+USER user
+
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH \
+    UVICORN_WS_PROTOCOL=websockets
+
+WORKDIR $HOME/app
+
+COPY --chown=user . $HOME/app
+
+RUN uv sync
+
+EXPOSE 7860
+
+CMD ["uv", "run", "chainlit", "run", "app.py", "--host", "0.0.0.0", "--port", "7860"]

README.md

@@ -1,10 +1,26 @@
----
-title: DEMO REFORMER
-emoji: 😻
-colorFrom: indigo
-colorTo: yellow
-sdk: docker
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+---
+title : Pilates App Fine_Tuned
+emoji: 📚
+colorFrom: red
+colorTo: red
+sdk: docker
+app_file: app.py
+pinned: false
+license: mit
+short_description: Tool to provide users reformer exercises
+startup_duration_timeout: 1h
+---
+
+# Pilates Reformer RAG App Fine_Tuned
+
+This Chainlit app answers questions using Pilates reformer videos and textbooks. All data is preloaded from `combined_data.json`.
+
+## Run Locally
+
+```bash
+uv run chainlit run app.py
+```
+
+## Or Deploy to Hugging Face Space with Docker
+Just upload this directory and you're done.
+

app.py

@@ -0,0 +1,112 @@
+import os
+import json
+from langchain_core.documents import Document
+from langchain_text_splitters import RecursiveCharacterTextSplitter
+from langchain_community.vectorstores import FAISS
+from langchain_huggingface import HuggingFaceEmbeddings
+from langchain_openai import ChatOpenAI
+from langchain.chains import RetrievalQA
+import chainlit as cl
+import stability_sdk.interfaces.gooseai.generation.generation_pb2 as generation
+from stability_sdk import client
+import io
+from PIL import Image
+
+# === Load and prepare data ===
+with open("combined_data.json", "r") as f:
+    raw_data = json.load(f)
+
+all_docs = [
+    Document(page_content=entry["content"], metadata=entry["metadata"])
+    for entry in raw_data
+]
+
+# === Split documents into chunks ===
+splitter = RecursiveCharacterTextSplitter(chunk_size=800, chunk_overlap=50)
+chunked_docs = splitter.split_documents(all_docs)
+
+# === Use your fine-tuned Hugging Face embeddings ===
+embedding_model = HuggingFaceEmbeddings(
+    model_name="AneetaXavier/reformer-pilates-embed-ft-49fc1835-9968-433d-9c45-1538ea91dcc9"
+)
+
+# === Set up FAISS vector store ===
+vectorstore = FAISS.from_documents(chunked_docs, embedding_model)
+retriever = vectorstore.as_retriever(search_kwargs={"k": 5})
+
+# === Load LLM ===
+llm = ChatOpenAI(model_name="gpt-4.1-mini", temperature=0)
+qa_chain = RetrievalQA.from_chain_type(llm=llm, retriever=retriever)
+
+# === Initialize Stability AI client ===
+stability_api = client.StabilityInference(
+    key=os.getenv('STABILITY_KEY'),
+    verbose=True,
+)
+
+# === Chainlit start event ===
+@cl.on_chat_start
+async def start():
+    await cl.Message(content = 
+"""👋 Welcome to your Reformer Pilates AI!
+
+Here's what you can do:
+• Ask questions about Reformer Pilates
+• Get individualized workouts based on your level, goals, and equipment
+• Get instant exercise modifications based on injuries or limitations
+
+Let's get started! 🚀""").send()
+    cl.user_session.set("qa_chain", qa_chain)
+
+# === Chainlit message handler ===
+@cl.on_message
+async def handle_message(message: cl.Message):
+    if message.content.lower().startswith('create an image'):
+        try:
+            # Extract the prompt from the message
+            prompt = message.content[15:].strip()
+            
+            # Generate the image
+            answers = stability_api.generate(
+                prompt=prompt,
+                seed=123,
+                steps=30,
+                cfg_scale=7.0,
+                width=512,
+                height=512,
+                samples=1,
+            )
+            
+            # Process the generated image
+            for resp in answers:
+                for artifact in resp.artifacts:
+                    if artifact.type == generation.ARTIFACT_IMAGE:
+                        # Convert the image data to a PIL Image
+                        img = Image.open(io.BytesIO(artifact.binary))
+                        
+                        # Save the image temporarily
+                        img_path = "temp_generated_image.png"
+                        img.save(img_path)
+                        
+                        # Send the image to the chat
+                        await cl.Message(
+                            content="Here's your generated image:",
+                            elements=[cl.Image(path=img_path, name="Generated Image")]
+                        ).send()
+                        
+                        # Clean up the temporary file
+                        os.remove(img_path)
+                        return
+            
+        except Exception as e:
+            await cl.Message(content=f"⚠️ Error generating image: {str(e)}").send()
+            return
+    
+    # Handle regular QA queries
+    chain = cl.user_session.get("qa_chain")
+    if chain:
+        try:
+            response = chain.run(message.content)
+        except Exception as e:
+            response = f"⚠️ Error: {str(e)}"
+        await cl.Message(response).send()

chainlit.md

@@ -0,0 +1,14 @@
+# Welcome to Chainlit! 🚀🤖
+
+Hi there, Developer! 👋 We're excited to have you on board. Chainlit is a powerful tool designed to help you prototype, debug and share applications built on top of LLMs.
+
+## Useful Links 🔗
+
+- **Documentation:** Get started with our comprehensive [Chainlit Documentation](https://docs.chainlit.io) 📚
+- **Discord Community:** Join our friendly [Chainlit Discord](https://discord.gg/k73SQ3FyUh) to ask questions, share your projects, and connect with other developers! 💬
+
+We can't wait to see what you create with Chainlit! Happy coding! 💻😊
+
+## Welcome screen
+
+To modify the welcome screen, edit the `chainlit.md` file at the root of your project. If you do not want a welcome screen, just leave this file empty.

pyproject.toml

@@ -0,0 +1,20 @@
+[project]
+name = "pilates_fine_tuned"
+requires-python = ">=3.13"
+version = "0.1.0"
+description = "A fine-tuned pilates project."
+dependencies = [
+    "langchain-huggingface>=0.0.6",
+    "chainlit>=2.5.5",
+    "faiss-cpu>=1.11.0",
+    "langchain>=0.3.25",
+    "langchain-community>=0.3.24",
+    "langchain-openai>=0.3.16",
+    "langchain-core>=0.0.1",  # Added langchain-core
+    "pymupdf>=1.25.5",
+    "pytube>=15.0.0",
+    "unstructured>=0.17.2",
+    "youtube-transcript-api>=1.0.3",
+    "websockets==11.0.3"
+]
+