数据集划分处理

#* coding:utf-8 *
import random

train = [] # 训练数据集
test = [] # 测试数据集
train_val = [] # 验证数据集

while (len(train)<41):
x = random.randint(63,113)
if (x not in train):
train.append(x)

while (len(test)<5):
x = random.randint(63,113)
if (x not in train) and (x not in test):
test.append(x)

while (len(train_val)<5):
x = random.randint(63,113)
if (x not in train) and ((x not in test)) and (x not in train_val):
train_val.append(x)

train_str = [] # 训练数据集name修改
with open(“F://business_license_data/train.txt”,‘w’) as f:
for a in range(len(train)):
if len(str(train[a])) == 1:
train_str.append(“0000” + str(train[a]))
if len(str(train[a])) == 2:
train_str.append(“000” + str(train[a]))
if len(str(train[a])) == 3:
train_str.append(“00” + str(train[a]))
if len(str(train[a])) == 4:
train_str.append(“0” + str(train[a]))
if len(str(train[a])) == 5:
train_str.append(str(train[a]))
f.write(train_str[a]+“\n”)

test_str = [] # 测试数据集name修改
with open(“F://business_license_data/test.txt”,‘w’) as f:
for a in range(len(test)):
if len(str(test[a])) == 1:
test_str.append(“0000” + str(test[a]))
if len(str(test[a])) == 2:
test_str.append(“000” + str(test[a]))
if len(str(test[a])) == 3:
test_str.append(“00” + str(test[a]))
if len(str(test[a])) == 4:
test_str.append(“0” + str(test[a]))
if len(str(test[a])) == 5:
test_str.append(str(test[a]))

    f.write(test_str[a] + "\n")

train_val_str = [] # 验证数据集name修改
with open(“F://business_license_data/trainval.txt”,‘w’) as f:
for a in range(len(train_val)):
if len(str(train_val[a])) == 1:
train_val_str.append(“0000” + str(train_val[a]))
if len(str(train_val[a])) == 2:
train_val_str.append(“000” + str(train_val[a]))
if len(str(train_val[a])) == 3:
train_val_str.append(“00” + str(train_val[a]))
if len(str(train_val[a])) == 4:
train_val_str.append(“0” + str(train_val[a]))
if len(str(train_val[a])) == 5:
train_val_str.append(str(train_val[a]))

    f.write(train_val_str[a] + "\n")