from huggingface_hub import snapshot_download

repo_id = "syn-nomos/EUR_Lex_Sum"

# Define patterns to include only the relevant data folder
allow_patterns = ["eudocsV6/**"]

print(f"Downloading EUR_Lex_Sum data from {repo_id}...")

local_dir = snapshot_download(
    repo_id=repo_id,
    repo_type="dataset",
    allow_patterns=allow_patterns,
    local_dir="./greek_eur_lex_sum",  # Local destination folder
    resume_download=True
)

print(f"Download complete! Data is located in: {local_dir}")