main.py

import logging

logging.basicConfig(
    filename="output/log_file.log",
    encoding="utf-8",
    format="%(asctime)s | %(levelname)s | %(module)s | %(message)s",
    level=logging.DEBUG,
)

logger = logging.getLogger(__name__)

logger.debug("===============  Started!  ===============")

from post_processing import run, generate_average_result, compare_results
from cutoff_eval import cutoff_eval
from load_data import load_train_test
from Models.Baseline import Baseline
from Models.ModifiedBaseline import ModifiedBaseline
from Models.DeepBaseline import DeepBaseline
from Models.DepthWiseBaseline import DepthWiseBaseline
from Models.AcousticModel import AcousticModel
from Models.GenreModel import GenreModel
from Models.ResNet1D import ResNet1D


model_dict = {
    "baseline": Baseline(),
    "modifiedbaseline": ModifiedBaseline(),
    "deepbaseline": DeepBaseline(),
    "depthwisebaseline": DepthWiseBaseline(),
    "acoustic": AcousticModel(),
    "genre": GenreModel(),
    "resnet": ResNet1D(),
}

model_names = list(model_dict.keys())


total_run = 10

## MAIN FUNCTION
# run(model_dict, total_run, use_long=True)

## ADD LONG AVERAGE
# model_long_names = [
#     "baseline_long",
#     "modifiedbaseline_long",
#     "deepbaseline_long",
#     "depthwisebaseline_long",
#     "acoustic_long",
#     "genre_long",
#     "resnet_long",
# ]

# for name in model_long_names:
#     generate_average_result(name)

## COMPARE RESULTS
# compare_results()


## CUTOFF EVAL
# cutoff_eval("baseline")
# cutoff_eval("genre")
# cutoff_eval("resnet")

## SONG NAME ID MAPPING
load_train_test(use_long=False)