Question

我正在尝试从输入生成器函数创建批处理列表，但它不会产生我期望的列表。

def batch_generator(items, batch_size):
new = []
i = 0

for item in items: 
    new.append(item)
    i += 1
    print('new: ', new, i)
    if i == batch_size:
        print('i = batch')
        i = 0
        yield new
        new = []


def _test_items_generator():
    for i in range(10):
        yield i

print(list(map(lambda x: len(x), 
               batch_generator(_test_items_generator(), 3))))

我试图让输出为[[0,1,2]，[3,4,5]，[6,7,8]，[9]] yield似乎是发送batch_size而不是新列表中的信息。试着让我理解这些发电机是如何工作的！

Answer 1

我认为问题在于你的最后一行：

print(list(map(lambda x: len(x), 
           batch_generator(_test_items_generator(), 3))))

batch_generator会产生new，其中包含一个列表。然后，您的map(lambda x: len(x)将返回每个列表的len。然后打印map()返回的长度列表。

以下是产生预期输出的代码：

def batch_generator(items, batch_size):
    new = []
    i = 0

    for item in items: 
        new.append(item)
        i += 1
        print('new: ', new, i)
        if i == batch_size:
            print('i = batch')
            i = 0
            yield new
            new = []

    yield new # yield the last list even if it is smaller than batch size

def _test_items_generator():
    for i in range(10):
        yield i

print(list( batch_generator(_test_items_generator(), 3)))

Answer 2

您的生成器工作正常。但在测试中，您将结果列表映射到其大小lambda x: len(x)

Answer 3

另一种batch_generator函数的方法：

def batch_generator(items, batch_size):
    current_batch = []

    for i, item in enumerate(items):
      current_batch.append(item)

      if len(current_batch) == batch_size:
        yield current_batch
        current_batch = []

    if len(current_batch) < batch_size:
      yield current_batch

python生成器创建从另一个genertator喂养的项目列表

3 个答案: