from experiment.sentence_wer_processor.spacy_pos_sentence_wer_processor import SpacyPosSentenceWerProcessor
from new_experiment.new_dependency_provider import get_experiment_repository, get_minio_audio_record_repository
from new_experiment.utils.get_spacy_model_name import get_spacy_model_name
from new_experiment.utils.loaded_remote_dataset_helper import LoadedRemoteDatasetHelper
from new_experiment.utils.property_helper import PropertyHelper
from sziszapangma.integration.experiment_manager import ExperimentManager


def run_spacy_pos_wer_pipeline(dataset_name: str, asr_name: str):
    repository = get_experiment_repository(dataset_name)
    record_provider = LoadedRemoteDatasetHelper(repository, get_minio_audio_record_repository(), dataset_name)
    language_code = dataset_name[:2]
    model_name = get_spacy_model_name(language_code)
    experiment_processor = ExperimentManager(
        record_id_iterator=record_provider,
        processing_tasks=[
            SpacyPosSentenceWerProcessor(
                model_name=model_name,
                gold_transcript_property_name=PropertyHelper.get_gold_transcript_raw(),
                asr_property_name=PropertyHelper.asr_result(asr_name),
                alignment_property_name=PropertyHelper.pos_alignment(asr_name, model_name),
                wer_property_name=PropertyHelper.pos_metrics(asr_name, model_name),
                task_name=f'SpacyPosSentenceWerProcessor___{dataset_name}___{asr_name}',
                require_update=False
            )
        ],
        experiment_repository=repository,
    )
    experiment_processor.process()


if __name__ == '__main__':
    run_spacy_pos_wer_pipeline('de_minds14', 'whisper_tiny')