harfoush4/arabic-eou-dataset

--- dataset_info: features: - name: text dtype: string - name: eou dtype: int64 - name: source dtype: string - name: word_position dtype: int64 - name: total_words dtype: int64 - name: is_final dtype: bool - name: word_count dtype: int64 - name: char_count dtype: int64 - name: position_ratio dtype: float64 - name: has_punctuation dtype: bool - name: ends_with_period dtype: bool - name: ends_with_question dtype: bool - name: ends_with_exclamation dtype: bool - name: is_single_backchannel dtype: bool - name: ends_with_continuation dtype: bool - name: starts_with_question dtype: bool - name: has_strong_eou dtype: bool - name: has_saudi_marker dtype: bool - name: is_very_short dtype: bool - name: is_short dtype: bool - name: is_medium dtype: bool - name: is_long dtype: bool splits: - name: train num_bytes: 3782488 num_examples: 35739 - name: validation num_bytes: 811094 num_examples: 7659 - name: test num_bytes: 810530 num_examples: 7659 download_size: 709554 dataset_size: 5404112 configs: - config_name: default data_files: - split: train path: data/train-* - split: validation path: data/validation-* - split: test path: data/test-* ---