from OpenGL.GL import * from gui.constants import StatisticLink from gui.ui_window import OptionGui from opengl_helper.screenshot import create_screenshot from processing.network_processing import NetworkProcessor from utility.file import FileHandler from utility.log_handling import setup_logger from utility.performance import track_time from utility.types import CameraPose from utility.window import WindowHandler, Window global options_gui options_gui = OptionGui() setup_logger("tool") def compute_render(some_name: str): global options_gui width, height = 1920, 1200 FileHandler().read_statistics() window_handler: WindowHandler = WindowHandler() window: Window = window_handler.create_window() window.set_callbacks() window.activate() logging.info(
os.path.abspath( os.path.join(os.path.dirname(sys.modules[__name__].__file__), ".."))) from typing import List from data.mnist_data_handler import split_mnist_data from data.model_data import ModelData, ModelTrainType from definitions import DATA_PATH from neural_network_preprocessing.create_mnist_model import create from neural_network_preprocessing.importance import ImportanceType, get_importance_type_name from neural_network_preprocessing.neural_network import ProcessedNetwork from processing.processing_handler import RecordingProcessingHandler from utility.log_handling import setup_logger from utility.recording_config import RecordingConfig setup_logger("sample_processing") # -------------------------------------------------change these settings-----------------------------------------------# name: str = "default" class_selection: List[int] or None = None # [0, 1, 2, 3, 4] importance_type: ImportanceType = ImportanceType(ImportanceType.GAMMA | ImportanceType.L1) basic_model_data: ModelData = create(name=name, batch_size=128, epochs=15, layer_data=[81, 49], regularized=False, class_selection=class_selection) # ---------------------------------------------------------------------------------------------------------------------#
def evaluate_importance(model_data: ModelData, importance_type: ImportanceType, importance_calculation: ImportanceCalculation): model_data.reload_model() importance_handler: ImportanceEvaluator = ImportanceEvaluator(model_data) importance_handler.setup(importance_type, importance_calculation) (x_train, y_train), (x_test, y_test), input_shape, num_classes = get_prepared_data( model_data.get_class_selection()) importance_handler.set_train_and_test_data(x_train, y_train, x_test, y_test) importance_handler.create_evaluation_data(10) setup_logger("evaluation") logging.info("Evaluation will take some time...") setup_plot() name: str = "default_all" layer_data: List[int] = [81, 49] model_data: ModelData = create(name=name, batch_size=128, epochs=15, layer_data=layer_data, regularized=False) split_suffix: str = "" if not os.path.exists("%smnist/mnist_train_split%s" % (DATA_PATH, split_suffix)) or not os.path.exists( "%smnist/mnist_test_split" % DATA_PATH):
from data.mnist_data_handler import get_prepared_data from data.model_data import ModelData from evaluation.evaluator import ImportanceEvaluator from utility.log_handling import setup_logger setup_logger("sample_evaluation") name: str = "default_all" model_data: ModelData = ModelData(name) model_data.reload_model() importance_handler: ImportanceEvaluator = ImportanceEvaluator(model_data) importance_handler.setup() (x_train, y_train), (x_test, y_test), input_shape, num_classes = get_prepared_data(model_data.get_class_selection()) importance_handler.set_train_and_test_data(x_train, y_train, x_test, y_test) importance_handler.create_evaluation_data(10)