把各個類別的原始數據打亂之后分成訓練集和測試集

但是我找不到 打亂 的代碼了 這版應該是按順序分成訓練集和測試集，并沒有把原始數據打亂

# -*- coding:utf-8 -*- # In[1]: import os import shutil import numpy as npbase_dir = r'C:\Users\10133\Desktop\bishe\matlab\traintest' if os.path.exists(base_dir):shutil.rmtree(base_dir) os.mkdir(base_dir) # 在該路徑下創建目錄train_dir = os.path.join(base_dir, 'train') # 訓練文件夾 os.mkdir(train_dir) test_dir = os.path.join(base_dir, 'test') # 測試文件夾 os.mkdir(test_dir)print('主目錄已經建立好了！')# In[2]: train_normal_dir = os.path.join(train_dir, '0') os.mkdir(train_normal_dir)train_fault1_dir = os.path.join(train_dir, '1') os.mkdir(train_fault1_dir)train_fault2_dir = os.path.join(train_dir, '2') os.mkdir(train_fault2_dir)train_fault3_dir = os.path.join(train_dir, '3') os.mkdir(train_fault3_dir)train_fault4_dir = os.path.join(train_dir, '4') os.mkdir(train_fault4_dir)train_fault5_dir = os.path.join(train_dir, '5') os.mkdir(train_fault5_dir)test_normal_dir = os.path.join(test_dir, '0') os.mkdir(test_normal_dir)test_fault1_dir = os.path.join(test_dir, '1') os.mkdir(test_fault1_dir)test_fault2_dir = os.path.join(test_dir, '2') os.mkdir(test_fault2_dir)test_fault3_dir = os.path.join(test_dir, '3') os.mkdir(test_fault3_dir)test_fault4_dir = os.path.join(test_dir, '4') os.mkdir(test_fault4_dir)test_fault5_dir = os.path.join(test_dir, '5') os.mkdir(test_fault5_dir)print('類別目錄已經建立好了！')train_normal = r'C:\Users\10133\Desktop\bishe\matlab\classification\0' train_fault1 = r'C:\Users\10133\Desktop\bishe\matlab\classification\1' train_fault2 = r'C:\Users\10133\Desktop\bishe\matlab\classification\2' train_fault3 = r'C:\Users\10133\Desktop\bishe\matlab\classification\3' train_fault4 = r'C:\Users\10133\Desktop\bishe\matlab\classification\4' train_fault5 = r'C:\Users\10133\Desktop\bishe\matlab\classification\5'original_dataset = r'C:\Users\10133\Desktop\bishe\matlab\SC4001jpg'#datapath=r'C:\Users\10133\Desktop\bishe\matlab\operation\sc4002e0_data.txt' labelpath=r'C:\Users\10133\PycharmProjects\practice\結果.txt'#x=np.loadtxt(fname=datapath,delimiter='\n') y=np.loadtxt(fname=labelpath,delimiter='\n')#y=[int(s) for s in y] #y=np.delete(y, -1, axis=0) #y=np.delete(y, 936, axis=0) print(y[0]) print(y.shape) print('finished!')for i in range(len(y)):if y[i]==0:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_normal, i, '.jpg')shutil.copyfile(src, dst)elif y[i]==1:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_fault1, i, '.jpg')shutil.copyfile(src, dst)elif y[i]==2:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_fault2, i, '.jpg')shutil.copyfile(src, dst)elif y[i]==3:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_fault3, i, '.jpg')shutil.copyfile(src, dst)elif y[i]==4:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_fault4, i, '.jpg')shutil.copyfile(src, dst)else:src = os.path.join(original_dataset, i, '.jpg')dst = os.path.join(train_fault5, i, '.jpg')shutil.copyfile(src, dst)

總結

以上是生活随笔為你收集整理的【3】基于深度神经网络的脑电睡眠分期方法研究（数据集分类）的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。