version 1.0 ready to merge

README.md

@@ -98,3 +98,4 @@ optional arguments:
 |v0.2|1.更新了滤波算法<br>2.修改了一些小bug<br>3.对部分函数进行解耦|
 |v0.3|1.训练了Unet模型<br>2.增加了部分命令行工具<br>4.重新设计目录结构|
 |v0.4|更新了mask计算方法，避免将信号计算为背景|
+|v1.0|完善第一阶段任务|
\ No newline at end of file

UnetBasedMethod.py

@@ -20,12 +20,9 @@ from utils.predict import predict_img
 from resCalc import save_img, get_subarea_info
 
 
-def divide_list(list, process_num):
-    res = []
-    step = math.ceil(len(list) / process_num)
-    for i in range(0, len(list), step):
-        res.append(list[i:i + step])
-    return res
+def divide_list(list, step):
+    n = math.ceil(len(list)/step)
+    return [list[i:i + n] for i in range(0, len(list), n)]
 
 
 def step_1(net, args, device, list, position):
@@ -91,7 +88,7 @@ if __name__ == '__main__':
     path = [(y, x) for y in filter(lambda x: x != '.DS_Store', os.listdir('data/imgs')) for x in filter(
         lambda x: x.endswith('.tif') and not x.endswith('dc.tif') and not x.endswith('DC.tif') and not x.endswith(
             'dc .tif'), os.listdir('data/imgs/' + y))]
-    seperate_path = divide_list(path, args.process)
+    seperate_path = divide_list(path,args.process)
 
     if args.step == 1:
         net = UNet(n_channels = 1, n_classes = 1)
@@ -110,7 +107,7 @@ if __name__ == '__main__':
 
     elif args.step == 2:
         dir = [x for x in filter(lambda x: x != '.DS_Store', os.listdir('data/imgs/'))]
-        sep_dir = divide_list(dir, args.process)
+        sep_dir = divide_list(dir,args.process)
 
         pool = Pool(args.process)
         for i, list in enumerate(sep_dir):

requirement.txt

@@ -5,3 +5,5 @@ matplotlib
 numpy
 opencv-python
 seaborn
+torch
+torchvision
\ No newline at end of file

resCalc.py

@@ -11,7 +11,7 @@ import re
 
 def save_img(img_list, dir, name):
     num = len(img_list)
-    plt.figure(figsize = (100, 20))
+    plt.figure(dpi = 300)
     plt.suptitle(name)
 
     for i, title in zip(range(num), img_list):