|
|
import os |
|
|
import sys |
|
|
from pathlib import Path |
|
|
|
|
|
from dotenv import load_dotenv |
|
|
import logging |
|
|
|
|
|
|
|
|
load_dotenv() |
|
|
|
|
|
|
|
|
logging.getLogger("httpx").setLevel(logging.WARNING) |
|
|
|
|
|
|
|
|
DOCS_FOLDER = os.getenv("DOCS_FOLDER", f"{Path.cwd()}/tmp") |
|
|
SUMMARIES_OUTPUT_DIR = os.path.join("/tmp", "summaries") |
|
|
|
|
|
|
|
|
os.makedirs(SUMMARIES_OUTPUT_DIR, exist_ok=True) |
|
|
|
|
|
|
|
|
EMBEDDING_MODEL = os.getenv("EMBEDDING_MODEL", "embed-v4.0") |
|
|
RERANKER_MODEL = os.getenv('RERANKER_MODEL', 'rerank-v3.5') |
|
|
LLM_MODEL = os.getenv("LLM_MODEL", "command-a-03-2025") |
|
|
|
|
|
|
|
|
CHUNK_SIZE = int(os.getenv("CHUNK_SIZE", "1000")) |
|
|
CHUNK_OVERLAP = int(os.getenv("CHUNK_OVERLAP", "100")) |
|
|
COHERERANK_TOPN = int(os.getenv("COHERERANK_TOPN", "100")) |
|
|
VECTOSTORE_TOPK = int(os.getenv("VECTOSTORE_TOPK", "100")) |