from experiments.experiment import experiments from Image_Process.Image_Generator import Image_generator from Training_Tools.Tools import Tool from model_data_processing.processing import make_label_list, Read_Image_Root_And_Image_Enhance from Load_process.LoadData import Load_Data_Prepare from Calculate_Process.Calculate import Calculate from merge_class.merge import merge from model_data_processing.processing_for_cut_image import Cut_Indepentend_Data from Load_process.LoadData import Loding_Data_Root from Load_process.file_processing import Process_File from utils.Stomach_Config import Training_Config, Loading_Config from Image_Process.Image_Mask_Ground_Truth_Processing import XMLAnnotationProcessor import time import torch if __name__ == "__main__": # 測試GPU是否可用 flag = torch.cuda.is_available() if not flag: print("CUDA不可用\n") else: print(f"CUDA可用，數量為{torch.cuda.device_count()}\n") # 测试GPU是否可用 device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') print(f"使用设备: {device}") print(f"GPU: {torch.cuda.get_device_name(0)}") tool = Tool() Status = 1 # 取得One-hot encording 的資料 tool.Set_OneHotEncording(Loading_Config["Training_Labels"]) Encording_Label = tool.Get_OneHot_Encording_Label() Label_Length = len(Loading_Config["Training_Labels"]) Prepare = Load_Data_Prepare() Indepentend = Cut_Indepentend_Data(Loading_Config["Train_Data_Root"], Loading_Config["Training_Labels"]) Merge = merge() Calculate_Tool = Calculate() file = Process_File() Train_Size = 0 # 讀取資料 # 將測試資料獨立出來 test_size = 0.2 Indepentend.IndependentData_main(Loading_Config["Test_Data_Root"], test_size) # # 創建處理切割的部分 # if not file.JudgeRoot_MakeDir(Loading_Config['Annotation_Training_Root']) and not file.JudgeRoot_MakeDir(Loading_Config['Annotation_Testing_Root']): # processor_train = XMLAnnotationProcessor( # dataset_root = Loading_Config["Train_Data_Root"], # ) # processor_test = XMLAnnotationProcessor( # dataset_root = Loading_Config["Test_Data_Root"], # ) # # 設定自訂樣式（可選） # processor_train.set_drawing_style( # box_color=(0, 0, 255), # 紅色邊界框 # text_color=(255, 255, 255), # 白色文字 # box_thickness=3, # font_scale=0.7 # ) # processor_test.set_drawing_style( # box_color=(0, 0, 255), # 紅色邊界框 # text_color=(255, 255, 255), # 白色文字 # box_thickness=3, # font_scale=0.7 # ) # print("XML標註處理器已準備就绪") # for Label_Iamge_List in Loading_Config["Label_Image_Labels"]: # if Label_Iamge_List == "CA": # Label = "stomach_cancer_Crop" # else: # Label = "Have_Question_Crop" # training_results = processor_train.process_multiple_xml(f"../Label_Image/{Label_Iamge_List}", Loading_Config['Annotation_Training_Root'], Label) # testing_results = processor_test.process_multiple_xml(f"../Label_Image/{Label_Iamge_List}", Loading_Config['Annotation_Testing_Root'], Label) # else: # print("Training and Testing annoation is exist!!!!") # # 讀取切割完成後的檔案 # print("Mask Ground truth is Finished\n") # Mask_load = Loding_Data_Root(Loading_Config["XML_Loading_Label"], Loading_Config['Annotation_Training_Root'], "") # Mask_Data_Dict_Data = Mask_load.process_main(False) # Total_Size_Lists = [] # print("Mask資料集總數") # for label in Loading_Config["XML_Loading_Label"]: # Train_Size += len(Mask_Data_Dict_Data[label]) # Total_Size_Lists.append(len(Mask_Data_Dict_Data[label])) # print(f"Labels: {label}, 總數為: {len(Mask_Data_Dict_Data[label])}") # print("總共有 " + str(Train_Size) + " 筆資料") # 讀取原始資料集 load = Loding_Data_Root(Loading_Config["Training_Labels"], Loading_Config["Train_Data_Root"], Loading_Config["ImageGenerator_Data_Root"]) Data_Dict_Data = load.process_main(False) # # 製作資料增強資料 # if not file.Judge_File_Exist(Loading_Config['Image enhance processing save root']): # for label in Loading_Config["Training_Labels"]: # Read_Image_Root_And_Image_Enhance(Data_Dict_Data[label], f"{Loading_Config['Image enhance processing save root']}/{label}") # tmp_load = Loding_Data_Root(Loading_Config["Training_Labels"], Loading_Config['Image enhance processing save root'], Loading_Config["ImageGenerator_Data_Root"]) # Data_Dict_Data = tmp_load.process_main(False) Total_Size_List = [] print("前處理後資料集總數") for label in Loading_Config["Training_Labels"]: Train_Size += len(Data_Dict_Data[label]) Total_Size_List.append(len(Data_Dict_Data[label])) print(f"Labels: {label}, 總數為: {len(Data_Dict_Data[label])}") print("總共有 " + str(Train_Size) + " 筆資料") # 做出跟資料相同數量的Label Classes = [] i = 0 for encording in Encording_Label: Classes.append(make_label_list(Total_Size_List[i], encording)) i += 1 # 將資料做成Dict的資料型態 Prepare.Set_Final_Dict_Data(Loading_Config["Training_Labels"], Data_Dict_Data, Classes, Label_Length) Final_Dict_Data = Prepare.Get_Final_Data_Dict() keys = list(Final_Dict_Data.keys()) # Mask_Keys = list(Mask_Data_Dict_Data.keys()) Training_Data = Merge.merge_all_image_data(Final_Dict_Data[keys[0]], Final_Dict_Data[keys[1]]) # 將訓練資料合併成一個list for i in range(2, Label_Length): Training_Data = Merge.merge_all_image_data(Training_Data, Final_Dict_Data[keys[i]]) # 將訓練資料合併成一個list Training_Label = Merge.merge_all_image_data(Final_Dict_Data[keys[Label_Length]], Final_Dict_Data[keys[Label_Length + 1]]) #將訓練資料的label合併成一個label的list for i in range(Label_Length + 2, 2 * Label_Length): Training_Label = Merge.merge_all_image_data(Training_Label, Final_Dict_Data[keys[i]]) # 將訓練資料合併成一個list # Training_Mask_Data = Merge.merge_all_image_data(Mask_Data_Dict_Data[Mask_Keys[0]], Mask_Data_Dict_Data[Mask_Keys[1]]) # 將訓練資料合併成一個list # for i in range(2, len(Mask_Keys)): # Training_Mask_Data = Merge.merge_all_image_data(Training_Mask_Data, Mask_Data_Dict_Data[Mask_Keys[i]]) # 將訓練資料合併成一個list experiment = experiments( Xception_Training_Data=Training_Data, Xception_Training_Label=Training_Label, Xception_Training_Mask_Data=None, status=Status ) start = time.time() experiment.processing_main() # 執行訓練方法 end = time.time() print(f"\n\n\n訓練時間：{end - start} 秒\n\n")