#!/usr/bin/python3
# -*- coding: utf-8 -*-
from collections import defaultdict
from datasets import load_dataset, DownloadMode


dataset_dict = load_dataset(
    "../../spam_detect.py",
    # name="enron_spam",
    # name="enron_spam_subset",
    # name="ling_spam",
    # name="sms_spam",
    # name="spam_assassin",
    # name="spam_detection",
    # name="sms_spam_collection",
    # name="spam_message",
    # name="spam_message_lr",
    name="trec07p",
    # name="youtube_spam_collection",
    split=None,
    cache_dir=None,
    download_mode=DownloadMode.FORCE_REDOWNLOAD
)


counter = defaultdict(int)
for split, dataset in dataset_dict.items():
    for sample in dataset:
        text = sample["text"]
        label = sample["label"]
        counter[label] += 1

count = "; ".join(sorted(["{}: {}".format(k, v) for k, v in counter.items()]))
print(count)


if __name__ == '__main__':
    pass