自学教程：Python产生batch数据的操作

import numpy as npdef batch_generator(all_data , batch_size, shuffle=True): """ :param all_data : all_data整个数据集，包含输入和输出标签 :param batch_size: batch_size表示每个batch的大小 :param shuffle: 是否打乱顺序 :return: """ # 输入all_datas的每一项必须是numpy数组，保证后面能按p所示取值 all_data = [np.array(d) for d in all_data] # 获取样本大小 data_size = all_data[0].shape[0] print("data_size: ", data_size) if shuffle:  # 随机生成打乱的索引  p = np.random.permutation(data_size)  # 重新组织数据  all_data = [d[p] for d in all_data] batch_count = 0 while True:  # 数据一轮循环(epoch)完成，打乱一次顺序  if batch_count * batch_size + batch_size > data_size:   batch_count = 0   if shuffle:    p = np.random.permutation(data_size)    all_data = [d[p] for d in all_data]  start = batch_count * batch_size  end = start + batch_size  batch_count += 1  yield [d[start: end] for d in all_data]

# 输入x表示有23个样本，每个样本有两个特征# 输出y表示有23个标签，每个标签取值为0或1x = np.random.random(size=[23, 2])y = np.random.randint(2, size=[23,1])count = x.shape[0]batch_size = 5epochs = 20batch_num = count // batch_sizebatch_gen = batch_generator([x, y], batch_size)for i in range(epochs): print("##### epoch %s ##### " % i) for j in range(batch_num):  batch_x, batch_y = next(batch_gen)  print("-----epoch=%s, batch=%s-----" % (i, j))  print(batch_x, batch_y)

import tensorflow as tfimport numpy as npdef _parse_function(x): num_list = np.arange(10) return num_listdef _from_tensor_slice(x): return tf.data.Dataset.from_tensor_slices(x)softmax_data = tf.data.Dataset.range(1000) # 构造一个队列softmax_data = softmax_data.map(lambda x:tf.py_func(_parse_function, [x], [tf.int32]))# 将数据进行传入softmax_data = softmax_data.flat_map(_from_tensor_slice) #将数据进行平铺, 将其变为一维的数据，from_tensor_slice将数据可以输出softmax_data = softmax_data.batch(1) #构造一个batch的数量softmax_iter = softmax_data.make_initializable_iterator() # 构造数据迭代器softmax_element = softmax_iter.get_next() # 获得一个batch的数据sess = tf.Session()sess.run(softmax_iter.initializer) # 数据迭代器的初始化操作print(sess.run(softmax_element)) # 实际获得一个数据print(sess.run(softmax_data))