import os
import pandas as pd

base_folder = "data/processed_large_prot_data"
in_csv = os.path.join(base_folder, "metadata.csv")
out_csv = in_csv  # overwrite in place; change if you prefer a new file

def update_path(path):
    name = os.path.basename(str(path))
    return os.path.join(base_folder, name)

df = pd.read_csv(in_csv)
df["processed_path"] = df["processed_path"].apply(update_path)
df.to_csv(out_csv, index=False)
print(f"Updated {len(df)} rows in {out_csv}")