claude-autopm 2.8.2 → 2.8.4

package/packages/plugin-ai/scripts/examples/langchain-rag-example.py

@@ -0,0 +1,366 @@
+#!/usr/bin/env python3
+"""
+LangChain RAG Example - Context7 Best Practices
+
+Demonstrates LangChain RAG patterns from Context7:
+- RunnablePassthrough.assign() for context injection
+- Vector store retrieval with MMR
+- RAG chain composition
+- Document chunking and embedding
+- Contextual compression for relevance
+
+Source: /langchain-ai/langchain (150 snippets, trust 9.2)
+"""
+
+import os
+import logging
+from typing import List
+from langchain_community.document_loaders import TextLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_openai import OpenAIEmbeddings, ChatOpenAI
+from langchain_community.vectorstores import Chroma
+from langchain_core.prompts import ChatPromptTemplate
+from langchain_core.runnables import RunnablePassthrough
+from langchain_core.output_parsers import StrOutputParser
+from operator import itemgetter
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Document Ingestion and Chunking
+# ===================================================================
+
+def create_sample_documents():
+    """Create sample documents for demonstration"""
+    documents = [
+        """
+        Python is a high-level, interpreted programming language known for its simplicity
+        and readability. It was created by Guido van Rossum and first released in 1991.
+        Python supports multiple programming paradigms including procedural, object-oriented,
+        and functional programming. It has a comprehensive standard library and a large
+        ecosystem of third-party packages available through PyPI.
+        """,
+        """
+        Machine Learning is a subset of artificial intelligence that focuses on the
+        development of algorithms that can learn from and make predictions on data.
+        Python is one of the most popular languages for machine learning due to libraries
+        like scikit-learn, TensorFlow, and PyTorch. These libraries provide implementations
+        of various ML algorithms including classification, regression, and clustering.
+        """,
+        """
+        LangChain is a framework for developing applications powered by language models.
+        It provides abstractions for working with LLMs, including chains, agents, and
+        retrievers. LangChain supports various vector stores for retrieval-augmented
+        generation (RAG), including Chroma, FAISS, and Pinecone. RAG combines the power
+        of language models with external knowledge bases for more accurate responses.
+        """,
+        """
+        Vector databases store high-dimensional vectors (embeddings) and enable efficient
+        similarity search. They are essential for RAG systems because they allow quick
+        retrieval of relevant documents based on semantic similarity. Popular vector
+        databases include Chroma (lightweight, local), Pinecone (cloud-based), and
+        FAISS (Facebook AI Similarity Search). Embeddings are typically generated
+        using models like OpenAI's text-embedding-3-small or sentence transformers.
+        """
+    ]
+
+    return documents
+
+
+def ingest_and_chunk_documents(texts: List[str]) -> List:
+    """
+    Ingest and chunk documents with Context7 best practices.
+
+    Context7 Pattern: RecursiveCharacterTextSplitter with optimal parameters
+    """
+    logger.info("Chunking documents...")
+
+    # Context7 recommended: RecursiveCharacterTextSplitter
+    text_splitter = RecursiveCharacterTextSplitter(
+        chunk_size=512,              # Optimal for embeddings
+        chunk_overlap=50,            # Maintain context across chunks
+        length_function=len,
+        separators=["\n\n", "\n", " ", ""]  # Hierarchical splitting
+    )
+
+    # Split documents
+    chunks = []
+    for i, text in enumerate(texts):
+        splits = text_splitter.split_text(text.strip())
+        for j, split in enumerate(splits):
+            chunks.append({
+                "page_content": split,
+                "metadata": {"source": f"doc_{i}", "chunk": j}
+            })
+
+    logger.info(f"✓ Created {len(chunks)} chunks from {len(texts)} documents")
+    return chunks
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Vector Store Creation
+# ===================================================================
+
+def create_vector_store(chunks: List[dict], persist_dir: str = "./chroma_db") -> Chroma:
+    """
+    Create vector store with Context7 best practices.
+
+    Context7 Pattern: Chroma with text-embedding-3-small
+    """
+    logger.info("Creating vector store...")
+
+    # Context7 recommended: text-embedding-3-small (cost-effective)
+    embeddings = OpenAIEmbeddings(model="text-embedding-3-small")
+
+    # Extract text and metadata
+    texts = [chunk["page_content"] for chunk in chunks]
+    metadatas = [chunk["metadata"] for chunk in chunks]
+
+    # Create Chroma vector store
+    vectorstore = Chroma.from_texts(
+        texts=texts,
+        embedding=embeddings,
+        metadatas=metadatas,
+        persist_directory=persist_dir
+    )
+
+    logger.info(f"✓ Vector store created with {len(texts)} documents")
+    return vectorstore
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Retriever Configuration
+# ===================================================================
+
+def create_mmr_retriever(vectorstore: Chroma, k: int = 4):
+    """
+    Create MMR retriever with Context7 best practices.
+
+    Context7 Pattern: MMR (Maximal Marginal Relevance) for diversity
+    """
+    logger.info("Creating MMR retriever...")
+
+    # Context7 recommended: MMR for diverse, relevant results
+    retriever = vectorstore.as_retriever(
+        search_type="mmr",
+        search_kwargs={
+            "k": k,                    # Number of documents to return
+            "fetch_k": k * 2,          # Fetch more candidates
+            "lambda_mult": 0.7         # Balance relevance (1.0) vs diversity (0.0)
+        }
+    )
+
+    logger.info(f"✓ MMR retriever created (k={k}, fetch_k={k*2}, lambda=0.7)")
+    return retriever
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: RAG Chain with RunnablePassthrough.assign()
+# ===================================================================
+
+def create_rag_chain(retriever):
+    """
+    Create RAG chain with Context7 best practices.
+
+    Context7 Pattern: RunnablePassthrough.assign() for elegant context handling
+    """
+    logger.info("Creating RAG chain...")
+
+    # Context7 recommended prompt template
+    template = """Answer the question based ONLY on the following context.
+If you cannot answer the question based on the context, say "I don't have enough information to answer."
+
+Context:
+{context}
+
+Question: {question}
+
+Answer:"""
+
+    prompt = ChatPromptTemplate.from_template(template)
+
+    # Context7: Use low temperature for factual responses
+    llm = ChatOpenAI(model="gpt-4", temperature=0)
+
+    # Context7 pattern: RunnablePassthrough.assign() for clean composition
+    rag_chain = (
+        RunnablePassthrough.assign(
+            context=itemgetter("question")
+                | retriever
+                | (lambda docs: "\n\n".join(doc.page_content for doc in docs))
+        )
+        | prompt
+        | llm
+        | StrOutputParser()
+    )
+
+    logger.info("✓ RAG chain created with RunnablePassthrough.assign()")
+    return rag_chain
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: RAG Query Examples
+# ===================================================================
+
+async def run_rag_queries(rag_chain):
+    """Run example RAG queries to demonstrate retrieval"""
+    print("\n" + "=" * 60)
+    print("RAG QUERY EXAMPLES")
+    print("=" * 60 + "\n")
+
+    queries = [
+        {
+            "title": "Query 1: Python Basics",
+            "question": "What is Python and who created it?"
+        },
+        {
+            "title": "Query 2: Machine Learning",
+            "question": "What Python libraries are popular for machine learning?"
+        },
+        {
+            "title": "Query 3: RAG and LangChain",
+            "question": "What is RAG and how does LangChain support it?"
+        },
+        {
+            "title": "Query 4: Unknown Topic (should return 'insufficient info')",
+            "question": "What is quantum computing?"
+        }
+    ]
+
+    for query in queries:
+        print(f"\n{query['title']}")
+        print("-" * 60)
+        print(f"Question: {query['question']}")
+        print()
+
+        try:
+            response = rag_chain.invoke({"question": query["question"]})
+            print(f"Answer: {response}")
+
+        except Exception as e:
+            logger.error(f"Query failed: {e}")
+            print(f"Error: {e}")
+
+        print()
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Retrieval Quality Testing
+# ===================================================================
+
+def test_retrieval_quality(retriever):
+    """Test retrieval quality with sample queries"""
+    print("\n" + "=" * 60)
+    print("RETRIEVAL QUALITY TEST")
+    print("=" * 60 + "\n")
+
+    test_queries = [
+        "Python programming language",
+        "machine learning libraries",
+        "vector databases and embeddings"
+    ]
+
+    for query in test_queries:
+        print(f"Query: '{query}'")
+        print("-" * 60)
+
+        # Retrieve documents
+        docs = retriever.get_relevant_documents(query)
+
+        for i, doc in enumerate(docs, 1):
+            preview = doc.page_content[:100].replace("\n", " ")
+            print(f"{i}. {preview}...")
+
+        print()
+
+
+# ===================================================================
+# MAIN DEMONSTRATION
+# ===================================================================
+
+def main():
+    """Run complete RAG demonstration"""
+    print("\n" + "=" * 60)
+    print("LangChain RAG Example - Context7 Best Practices")
+    print("=" * 60 + "\n")
+
+    try:
+        # Check API key
+        if not os.getenv("OPENAI_API_KEY"):
+            raise ValueError("OPENAI_API_KEY environment variable not set")
+
+        # 1. Create sample documents
+        print("Step 1: Creating sample documents...")
+        documents = create_sample_documents()
+        print(f"✓ Created {len(documents)} sample documents\n")
+
+        # 2. Chunk documents
+        print("Step 2: Chunking documents...")
+        chunks = ingest_and_chunk_documents(documents)
+        print()
+
+        # 3. Create vector store
+        print("Step 3: Creating vector store...")
+        vectorstore = create_vector_store(chunks)
+        print()
+
+        # 4. Create retriever
+        print("Step 4: Creating MMR retriever...")
+        retriever = create_mmr_retriever(vectorstore, k=3)
+        print()
+
+        # 5. Test retrieval quality
+        test_retrieval_quality(retriever)
+
+        # 6. Create RAG chain
+        print("Step 5: Creating RAG chain...")
+        rag_chain = create_rag_chain(retriever)
+        print()
+
+        # 7. Run RAG queries
+        import asyncio
+        asyncio.run(run_rag_queries(rag_chain))
+
+        # Summary
+        print("\n" + "=" * 60)
+        print("RAG EXAMPLE COMPLETED SUCCESSFULLY")
+        print("=" * 60)
+        print("\nContext7 Patterns Demonstrated:")
+        print("1. ✅ RecursiveCharacterTextSplitter for hierarchical chunking")
+        print("2. ✅ text-embedding-3-small for cost-effective embeddings")
+        print("3. ✅ MMR retrieval for diverse, relevant results")
+        print("4. ✅ RunnablePassthrough.assign() for clean composition")
+        print("5. ✅ Temperature=0 for factual RAG responses")
+        print("6. ✅ Context boundary enforcement in prompt")
+        print("\nSource: /langchain-ai/langchain (150 snippets, trust 9.2)")
+
+    except ValueError as e:
+        logger.error(f"Configuration error: {e}")
+        print(f"\n❌ Error: {e}")
+        print("\nPlease set OPENAI_API_KEY environment variable:")
+        print("export OPENAI_API_KEY='your-api-key-here'")
+
+    except Exception as e:
+        logger.error(f"Unexpected error: {e}")
+        print(f"\n❌ Unexpected error: {e}")
+        import traceback
+        traceback.print_exc()
+
+
+if __name__ == "__main__":
+    print("LangChain RAG Example - Context7 Best Practices")
+    print("=" * 60)
+    print("")
+    print("This example demonstrates Context7-verified patterns for:")
+    print("- Document chunking with RecursiveCharacterTextSplitter")
+    print("- Vector store creation with Chroma")
+    print("- MMR retrieval for diverse results")
+    print("- RAG chain composition with RunnablePassthrough.assign()")
+    print("- Context boundary enforcement")
+    print("")
+    print("Source: /langchain-ai/langchain (150 snippets, trust 9.2)")
+    print("")
+
+    main()

package/packages/plugin-ai/scripts/examples/mlflow-tracking-example.py

@@ -0,0 +1,224 @@
+#!/usr/bin/env python3
+"""
+MLflow Tracking Example - Context7 Best Practices
+
+Demonstrates MLflow patterns from Context7:
+- Experiment tracking and organization
+- Parameter and metric logging
+- Model registry and versioning
+- Artifact storage
+- Run comparison
+
+Source: /mlflow/mlflow (3,114 snippets, trust 9.1)
+"""
+
+import mlflow
+import mlflow.sklearn
+from mlflow.tracking import MlflowClient
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.datasets import load_iris
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import accuracy_score, f1_score
+import logging
+
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Experiment Organization
+# ===================================================================
+
+def setup_experiment(experiment_name: str = "iris-classification"):
+    """
+    Context7 Pattern: Organize runs into experiments
+    """
+    logger.info(f"Setting up experiment: {experiment_name}")
+
+    # Context7 pattern: Set experiment for organization
+    mlflow.set_experiment(experiment_name)
+
+    logger.info("✓ Experiment setup complete")
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Parameter and Metric Logging
+# ===================================================================
+
+def train_and_log_model(n_estimators: int, max_depth: int):
+    """
+    Context7 Pattern: Log params, metrics, and models
+    """
+    logger.info(f"Training model (n_estimators={n_estimators}, max_depth={max_depth})")
+
+    # Load data
+    iris = load_iris()
+    X_train, X_test, y_train, y_test = train_test_split(
+        iris.data, iris.target, test_size=0.2, random_state=42
+    )
+
+    # Context7 pattern: Use mlflow.start_run() context manager
+    with mlflow.start_run():
+        # Log parameters
+        mlflow.log_param("n_estimators", n_estimators)
+        mlflow.log_param("max_depth", max_depth)
+        mlflow.log_param("random_state", 42)
+
+        # Train model
+        model = RandomForestClassifier(
+            n_estimators=n_estimators,
+            max_depth=max_depth,
+            random_state=42
+        )
+        model.fit(X_train, y_train)
+
+        # Make predictions
+        y_pred = model.predict(X_test)
+
+        # Calculate metrics
+        accuracy = accuracy_score(y_test, y_pred)
+        f1 = f1_score(y_test, y_pred, average="weighted")
+
+        # Log metrics
+        mlflow.log_metric("accuracy", accuracy)
+        mlflow.log_metric("f1_score", f1)
+
+        # Context7 pattern: Log model with signature
+        mlflow.sklearn.log_model(
+            model,
+            "model",
+            registered_model_name="iris-classifier"
+        )
+
+        # Log tags
+        mlflow.set_tag("model_type", "RandomForest")
+        mlflow.set_tag("dataset", "iris")
+
+        run_id = mlflow.active_run().info.run_id
+        logger.info(f"✓ Run complete: {run_id} (accuracy={accuracy:.3f})")
+
+        return run_id, accuracy, f1
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Run Comparison
+# ===================================================================
+
+def compare_runs():
+    """
+    Context7 Pattern: Compare multiple runs
+    """
+    logger.info("Comparing multiple model configurations...")
+
+    configs = [
+        {"n_estimators": 10, "max_depth": 3},
+        {"n_estimators": 50, "max_depth": 5},
+        {"n_estimators": 100, "max_depth": 10},
+    ]
+
+    results = []
+
+    for config in configs:
+        run_id, accuracy, f1 = train_and_log_model(**config)
+        results.append({
+            "run_id": run_id,
+            "config": config,
+            "accuracy": accuracy,
+            "f1": f1
+        })
+
+    logger.info("✓ Model comparison complete")
+    return results
+
+
+# ===================================================================
+# CONTEXT7 PATTERN: Model Registry
+# ===================================================================
+
+def get_best_model():
+    """
+    Context7 Pattern: Retrieve best model from registry
+    """
+    logger.info("Retrieving best model from registry...")
+
+    client = MlflowClient()
+
+    # Get all versions of registered model
+    model_name = "iris-classifier"
+    versions = client.search_model_versions(f"name='{model_name}'")
+
+    if not versions:
+        logger.warning("No registered models found")
+        return None
+
+    # Find version with highest accuracy (from tags/metrics)
+    best_version = None
+    best_accuracy = 0
+
+    for version in versions:
+        run = client.get_run(version.run_id)
+        accuracy = run.data.metrics.get("accuracy", 0)
+
+        if accuracy > best_accuracy:
+            best_accuracy = accuracy
+            best_version = version
+
+    if best_version:
+        logger.info(f"✓ Best model: v{best_version.version} (accuracy={best_accuracy:.3f})")
+
+    return best_version
+
+
+# ===================================================================
+# MAIN DEMONSTRATION
+# ===================================================================
+
+def main():
+    """Run complete MLflow tracking demonstration"""
+    print("\n" + "=" * 60)
+    print("MLflow Tracking Example - Context7 Best Practices")
+    print("=" * 60 + "\n")
+
+    # 1. Setup experiment
+    print("Step 1: Setting up experiment...")
+    setup_experiment()
+    print()
+
+    # 2. Compare multiple runs
+    print("Step 2: Training and comparing models...")
+    print("-" * 60)
+    results = compare_runs()
+    print()
+
+    # 3. Display results
+    print("Step 3: Run Comparison Results")
+    print("-" * 60)
+    for i, result in enumerate(results, 1):
+        config = result["config"]
+        print(f"{i}. n_estimators={config['n_estimators']}, max_depth={config['max_depth']}")
+        print(f"   Accuracy: {result['accuracy']:.3f}, F1: {result['f1']:.3f}")
+        print(f"   Run ID: {result['run_id'][:8]}...")
+    print()
+
+    # 4. Get best model
+    print("Step 4: Retrieving best model from registry...")
+    print("-" * 60)
+    best_model = get_best_model()
+    print()
+
+    # Summary
+    print("\n" + "=" * 60)
+    print("MLflow TRACKING COMPLETED SUCCESSFULLY")
+    print("=" * 60)
+    print("\nContext7 Patterns Demonstrated:")
+    print("1. ✅ Experiment organization with set_experiment()")
+    print("2. ✅ Parameter logging with log_param()")
+    print("3. ✅ Metric logging with log_metric()")
+    print("4. ✅ Model logging with sklearn.log_model()")
+    print("5. ✅ Model registry for versioning")
+    print("\nSource: /mlflow/mlflow (3,114 snippets, trust 9.1)")
+    print(f"\nView results: mlflow ui")
+
+
+if __name__ == "__main__":
+    main()