【发布时间】:2018-08-10 18:59:49
【问题描述】:
说,我们输入了x和标签y:
iterator = tf.data.Iterator.from_structure((x_type, y_type), (x_shape, y_shape))
tf_x, tf_y = iterator.get_next()
现在我使用 generate 函数来创建数据集:
def gen():
for ....: yield (x, y)
ds = tf.data.Dataset.from_generator(gen, (x_type, y_type), (x_shape, y_shape))
在我的图表中,我使用tf_x 和tf_y 进行训练,这很好。但现在我想做参考,我没有标签y。我提出的一种解决方法是伪造一个 y(如 tf.zeros(y_shape)),然后使用占位符来初始化迭代器。
x_placeholder = tf.placeholder(...)
y_placeholder = tf.placeholder(...)
ds = tf.data.Dataset.from_tensors((x_placeholder, y_placeholder))
ds_init_op = iterator.make_initializer(ds)
sess.run(ds_init_op, feed_dict={x_placeholder=x, y_placeholder=fake(y))})
我的问题是,有没有更清洁的方法来做到这一点?在推断期间没有伪造y?
更新:
我实验了一下,貌似少了一个数据集操作unzip:
import numpy as np
import tensorflow as tf
x_type = tf.float32
y_type = tf.float32
x_shape = tf.TensorShape([None, 128])
y_shape = tf.TensorShape([None, 10])
x_shape_nobatch = tf.TensorShape([128])
y_shape_nobatch = tf.TensorShape([10])
iterator_x = tf.data.Iterator.from_structure((x_type,), (x_shape,))
iterator_y = tf.data.Iterator.from_structure((y_type,), (y_shape,))
def gen1():
for i in range(100):
yield np.random.randn(128)
ds1 = tf.data.Dataset.from_generator(gen1, (x_type,), (x_shape_nobatch,))
ds1 = ds1.batch(5)
ds1_init_op = iterator_x.make_initializer(ds1)
def gen2():
for i in range(80):
yield np.random.randn(128), np.random.randn(10)
ds2 = tf.data.Dataset.from_generator(gen2, (x_type, y_type), (x_shape_nobatch, y_shape_nobatch))
ds2 = ds2.batch(10)
# my ds2 has two tensors in one element, now the problem is
# how can I unzip this dataset so that I can apply them to iterator_x and iterator_y?
# such as:
ds2_x, ds2_y = tf.data.Dataset.unzip(ds2) #?? missing this unzip operation!
ds2_x_init_op = iterator_x.make_initializer(ds2_x)
ds2_y_init_op = iterator_y.make_initializer(ds2_y)
tf_x = iterator_x.get_next()
tf_y = iterator_y.get_next()
【问题讨论】:
标签: python tensorflow dataset tensorflow-datasets