config.yaml

# Database Configuration
database:
  db_host: "localhost"         # Database host address
  db_port: 1972                # Database port number
  db_user: "SuperUser"         # Database username (default for docker)
  db_password: "SYS"           # Database password (default for docker ISC_DEFAULT_PASSWORD)
  db_namespace: "USER"         # Database namespace

# Default Embedding Model Configuration
embedding_model:
  name: "sentence-transformers/all-MiniLM-L6-v2"    # Name of the sentence transformer model
  dimension: 384               # Embedding dimension

# Default Chunking Parameters
chunking:
  chunk_size: 1000             # Target size of text chunks (e.g., in tokens or characters)
  chunk_overlap: 200           # Number of tokens/characters to overlap between chunks

# Data Directories
paths:
  data_dir: "data/"                   # Root directory for all data
  pmc_sample_dir: "data/pmc_sample/"  # Directory for PMC sample documents

# Logging Configuration
logging:
  log_level: "INFO"            # Logging level (e.g., DEBUG, INFO, WARNING, ERROR, CRITICAL)
  log_format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s" # Logging format string

# Test Configuration
testing:
  min_docs_e2e: 1000         # Minimum documents required for E2E tests

# Storage Backend Configuration for iris_rag
storage:
  backends:
    iris:
      type: "iris"
      connection_type: "dbapi"
      schema: "RAG"
      table_prefix: ""
      vector_dimension: 384

# Pipeline Configuration for iris_rag
pipelines:
  basic:
    chunk_size: 1000
    chunk_overlap: 200
    default_top_k: 5
    embedding_batch_size: 32
  crag:
    chunk_size: 1000
    chunk_overlap: 200
    default_top_k: 5

# Embedding Configuration for iris_rag
embeddings:
  backend: "sentence_transformers"
  model: "sentence-transformers/all-MiniLM-L6-v2"
  dimension: 384