import pandas as pd

# Extract sentences from one of the CSV files (let's use subnum.csv)
df = pd.read_csv("./probingData/hindi/subnum.csv")

# Save sentences to a text file
with open("./gold/hindi/sentences.txt", "w", encoding="utf-8") as f:
    for sentence in df["sentences"]:
        if pd.notna(sentence):  # Skip empty/NaN values
            f.write(sentence.strip() + "\n")

print("Sentences extracted successfully!")
print(f"Total sentences: {len(df)}")