InitConfig.py 2.6 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081
  1. import logging
  2. import os
  3. from chromadb.utils import embedding_functions
  4. from embedchain.config.BaseConfig import BaseConfig
  5. class InitConfig(BaseConfig):
  6. """
  7. Config to initialize an embedchain `App` instance.
  8. """
  9. def __init__(self, log_level=None, ef=None, db=None, host=None, port=None, id=None):
  10. """
  11. :param log_level: Optional. (String) Debug level
  12. ['DEBUG', 'INFO', 'WARNING', 'ERROR', 'CRITICAL'].
  13. :param ef: Optional. Embedding function to use.
  14. :param db: Optional. (Vector) database to use for embeddings.
  15. :param id: Optional. ID of the app. Document metadata will have this id.
  16. :param host: Optional. Hostname for the database server.
  17. :param port: Optional. Port for the database server.
  18. """
  19. self._setup_logging(log_level)
  20. if db is None:
  21. from embedchain.vectordb.chroma_db import ChromaDB
  22. self.db = ChromaDB(ef=ef)
  23. else:
  24. self.db = db
  25. self.ef = ef
  26. self.host = host
  27. self.port = port
  28. self.id = id
  29. return
  30. def _set_embedding_function(self, ef):
  31. self.ef = ef
  32. return
  33. def _set_embedding_function_to_default(self):
  34. """
  35. Sets embedding function to default (`text-embedding-ada-002`).
  36. :raises ValueError: If the template is not valid as template should contain
  37. $context and $query
  38. """
  39. if os.getenv("OPENAI_API_KEY") is None and os.getenv("OPENAI_ORGANIZATION") is None:
  40. raise ValueError("OPENAI_API_KEY or OPENAI_ORGANIZATION environment variables not provided") # noqa:E501
  41. self.ef = embedding_functions.OpenAIEmbeddingFunction(
  42. api_key=os.getenv("OPENAI_API_KEY"),
  43. organization_id=os.getenv("OPENAI_ORGANIZATION"),
  44. model_name="text-embedding-ada-002",
  45. )
  46. return
  47. def _set_db(self, db):
  48. if db:
  49. self.db = db
  50. return
  51. def _set_db_to_default(self):
  52. """
  53. Sets database to default (`ChromaDb`).
  54. """
  55. from embedchain.vectordb.chroma_db import ChromaDB
  56. self.db = ChromaDB(ef=self.ef, host=self.host, port=self.port)
  57. def _setup_logging(self, debug_level):
  58. level = logging.WARNING # Default level
  59. if debug_level is not None:
  60. level = getattr(logging, debug_level.upper(), None)
  61. if not isinstance(level, int):
  62. raise ValueError(f"Invalid log level: {debug_level}")
  63. logging.basicConfig(format="%(asctime)s [%(name)s] [%(levelname)s] %(message)s", level=level)
  64. self.logger = logging.getLogger(__name__)
  65. return