File size: 2,599 Bytes
6059138 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 | from __future__ import annotations
import os
from dataclasses import dataclass
from dotenv import load_dotenv
load_dotenv()
@dataclass(frozen=True)
class AppConfig:
mistral_api_key: str
memory_data_dir: str = "./data/faiss"
mistral_model: str = "mistral-small-latest"
embedding_model: str = "mistral-embed"
mistral_max_retries: int = 6
mistral_max_concurrent_requests: int = 1
mistral_timeout_seconds: int = 120
rate_limit_base_delay_seconds: float = 1.0
rate_limit_max_delay_seconds: float = 16.0
dense_weight: float = 0.55
sparse_weight: float = 0.45
top_k: int = 6
@classmethod
def from_env(cls) -> "AppConfig":
mistral_api_key = os.getenv("MISTRAL_API_KEY") or os.getenv("API_KEY")
missing = []
if not mistral_api_key:
missing.append("API_KEY or MISTRAL_API_KEY")
if missing:
formatted = ", ".join(missing)
raise ValueError(f"Missing required environment variable(s): {formatted}")
memory_data_dir = os.getenv("MEMORY_DATA_DIR", "./data/faiss")
mistral_model = os.getenv("MISTRAL_MODEL", "mistral-small-latest")
embedding_model = os.getenv("EMBEDDING_MODEL", "mistral-embed")
if embedding_model == "mistrall-embed":
embedding_model = "mistral-embed"
mistral_max_retries = int(os.getenv("MISTRAL_MAX_RETRIES", "6"))
mistral_max_concurrent_requests = int(os.getenv("MISTRAL_MAX_CONCURRENT_REQUESTS", "1"))
mistral_timeout_seconds = int(os.getenv("MISTRAL_TIMEOUT_SECONDS", "120"))
rate_limit_base_delay_seconds = float(os.getenv("RATE_LIMIT_BASE_DELAY_SECONDS", "1.0"))
rate_limit_max_delay_seconds = float(os.getenv("RATE_LIMIT_MAX_DELAY_SECONDS", "16.0"))
dense_weight = float(os.getenv("DENSE_WEIGHT", "0.55"))
sparse_weight = float(os.getenv("SPARSE_WEIGHT", "0.45"))
top_k = int(os.getenv("TOP_K", "6"))
return cls(
mistral_api_key=mistral_api_key,
memory_data_dir=memory_data_dir,
mistral_model=mistral_model,
embedding_model=embedding_model,
mistral_max_retries=mistral_max_retries,
mistral_max_concurrent_requests=mistral_max_concurrent_requests,
mistral_timeout_seconds=mistral_timeout_seconds,
rate_limit_base_delay_seconds=rate_limit_base_delay_seconds,
rate_limit_max_delay_seconds=rate_limit_max_delay_seconds,
dense_weight=dense_weight,
sparse_weight=sparse_weight,
top_k=top_k,
)
|