import numpy as np
import pickle
import json
import os
import sys

# add code/ to path so we can import our preprocessing_utils
sys.path.append('code')
from preprocessing_utils import preprocess_embeddings

# Config 
DATA_PATH = '/scratch/users/s214/lab3'

# the text/embeddings are the same for both subjects (same stories), but fMRI lengths
# differ slightly per subject so we preprocess separately per subject
SUBJECTS = ['subject2', 'subject3']

# short names used in output filenames following naming convention
SUBJECT_IDS = {'subject2': 's2', 'subject3': 's3'}

# where to save the processed embeddings
OUTPUT_DIR = 'data/embeddings'
os.makedirs(OUTPUT_DIR, exist_ok=True)