# Author: Simon-Pierre Boucher — contact@spboucher.ai
#
"""Central configuration: repository-relative paths and constants.

Every path is derived from the repository root so the pipeline runs from any
working directory. Override the database location with the WP2_DB_PATH
environment variable if the raw database lives elsewhere.
"""

import os
from pathlib import Path

ROOT = Path(__file__).resolve().parents[1]

DATA_RAW = ROOT / "data" / "raw"
DATA_PROCESSED = ROOT / "data" / "processed"
FIGURES_DIR = ROOT / "figures"
RESULTS_DIR = ROOT / "results"

DB_PATH = Path(os.environ.get("WP2_DB_PATH", DATA_RAW / "louka.db"))

# Intermediate artifacts (produced by the numbered scripts)
HOUSES_PARQUET = DATA_PROCESSED / "houses.parquet"
EMBEDDINGS_NPY = DATA_PROCESSED / "embeddings_maisons.npy"
SIM_MATRIX_NPY = DATA_PROCESSED / "sim_matrix_maisons.npy"
ANALYSIS_CSV = DATA_PROCESSED / "hedonic_maison_results.csv"

# Sentence-transformers model used for all embeddings (384 dimensions)
EMBEDDING_MODEL = "all-MiniLM-L6-v2"
ENCODE_BATCH_SIZE = 256

# Structural covariates of the hedonic model
STRUCTURAL_VARS = ["bedrooms", "bathrooms", "half_baths", "parking", "stories", "land_size"]

SEED = 42