whitekirin/Stomach_Cancer_Pytorch
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
9 Commits 3 Branches 0 Tags 261 MiB
Jupyter Notebook 93.5%
Python 6.5%
3f3fa57a02
Go to file
whitekirin 3f3fa57a02 20250226 Commits: Modification tqdm in the training step
2025-02-26 20:15:31 +08:00
_validation 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
all_models_tools 20250224 commits: Data Argumentation had done 5x, and validation Accuracy just only have 33% 2025-02-24 22:57:27 +08:00
application 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
Calculate_Process 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
draw_tools 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
experiments 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
Image_Process 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
Load_process 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
merge_class 20250224 commits: Data Argumentation had done 5x, and validation Accuracy just only have 33% 2025-02-24 22:57:27 +08:00
model_data_processing 20250218 commits: The programing can run and it's validation accuracy is 50 percent 2025-02-18 20:51:10 +08:00
Model_Loss 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
process_csv_file 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
Read_and_process_image 20250218 commits: The programing can run and it's validation accuracy is 50 percent 2025-02-18 20:51:10 +08:00
SCP_Process 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
To_load_Input_Data 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
Training_Tools 20250218 commits: The programing can run and it's validation accuracy is 50 percent 2025-02-18 20:51:10 +08:00
best_model( 2023-10-17 )-2.h5 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
claculate_output_data.py 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
main.py 20250226 Commits: Modification tqdm in the training step 2025-02-26 20:15:31 +08:00
Processing_image.py 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
README.md 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
test.ipynb 20250218 commits: The script can run but Grad-cam class have some question 2025-02-18 22:54:45 +08:00
testing_Labels_Accuracy.py 20250210 Modification Programing can run, and train,validation, and test still can training 2025-02-11 02:52:36 +08:00
testing.py 20250213 commit: To added list to Dataset Script 2025-02-13 03:11:54 +08:00

README.md

main.py: 主程式檔

load_process

負責讀取影像檔案、分割獨立資料(測試、驗證)、讀取獨立資料、一般檔案的操作

File_Process : 檔案操作的主程式，包含開檔、創立檔案、判斷檔案是否存在等都是他負責的範圍。是一般物件也是LoadData的父物件 LoadData : 讀檔主程式，一切讀檔動作由他開始。繼承File_Process(子物件) Cutting_Indepentend_Image : 讀取獨立資料(testing、Validation)的物件

Image_Process

負責進行資料擴增、影像處理等的操作

  • Generator_Content : 負責建立基礎Generator項目，為Image_Generator的父類別
  • Image_Generator : 負責製造資料擴增的資料，並將資料存到檔案中。繼承Generator_Content(子物件)
  • image_enhancement : 負責進行影像處理並將資料回傳

Model_Tools

負責進行模型的基礎架構，包含Convolution、Dense、以及其他模型的配件

  • All_Model_Tools : 所有模型的附加工具，是所有的父類別

    CNN

    包含所有CNN的工具與應用架構

    • CNN_Tools : 為卷積層的工具，包含一維、二維、三維捲積。CNN_Application的父類別，繼承All_Model_Tools(子類別)
    • CNN_Application : 為Convolution的應用架構。繼承CNN_Tools(子類別)

    Dense

    包含所有Dense的應用

    • Dense_Application : 為全連階層工具，包含一般Dense layer與增加正則化之Dense layer。繼承All_Model_Tools()

    Model_Construction

    包含所有要進行實驗的模型架構

    • Model_Constructions : 所有模型的實驗架構

Data_Merge

負責進行資料的合併

  • Merge : 負責合併Dict、List到List並匯出

initalization

負責初始化特定物件

  • Img_initalization : 針對影像資料的初始化
  • Data_Initalization : 針對數據資料的初始化

Validation_Program

負責驗證程式碼內的資料型態或輸入錯誤等問題

  • Validation : 驗證程式碼錯誤

Draw

負責畫圖的工具

  • Draw_Tools : 畫出混淆矩陣、走勢圖的工具
  • Grad_CAM : 畫出模型可視化的熱力圖的工具

Experiment

執行實驗的主程式

  • Experiment : 負責執行讀檔、設定模型Compile的細節、執行訓練、驗證結果等功能