增加mnsit数据集张量流

时间:2018-04-07 00:58:46

标签: python tensorflow machine-learning computer-vision mnist

我正在尝试扩充MNIST数据集。这是我试过的。无法取得任何成功。

from tensorflow.examples.tutorials.mnist import input_data
import tensorflow as tf

X = mnist.train.images
y = mnist.train.labels

def flip_images(X_imgs):
    X_flip = []
    tf.reset_default_graph()
    X = tf.placeholder(tf.float32, shape = (28, 28, 1))
    input_d = tf.reshape(X_imgs, [-1, 28, 28, 1])
    tf_img1 = tf.image.flip_left_right(X)
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        for img in input_d:
            flipped_imgs = sess.run([tf_img1], feed_dict = {X: img})
            X_flip.extend(flipped_imgs)
    X_flip = np.array(X_flip, dtype = np.float32)
    return X_flip

flip = flip_images(X)

我做错了什么?我似乎无法弄明白。

错误:

Line: for img in input_d:
raise TypeError("'Tensor' object is not iterable.")
TypeError: 'Tensor' object is not iterable

1 个答案:

答案 0 :(得分:2)

首先,请注意您的tf.reshape将类型从ndarray更改为张量。它将需要一个.eval()调用才能将其恢复。在for循环中,您尝试迭代张量(不是列表或真正的可迭代),考虑在数字上进行索引,如下所示:

X = mnist.train.images
y = mnist.train.labels

def flip_images(X_imgs):

    X_flip = []
    tf.reset_default_graph()
    X = tf.placeholder(tf.float32, shape = (28, 28, 1))

    input_d = tf.reshape(X_imgs, [-1, 28, 28, 1])
    tf_img1 = tf.image.flip_left_right(X)

    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())        
        for img_ind in range(input_d.shape[0]):
            img = input_d[img_ind].eval()
            flipped_imgs = sess.run([tf_img1], feed_dict={X: img})
            X_flip.extend(flipped_imgs)
    X_flip = np.array(X_flip, dtype = np.float32)
    return X_flip

flip = flip_images(X)

如果这可以解决您的问题,请告诉我们!可能希望将范围设置为一个小常量进行测试,如果你没有GPU,这可能需要一段时间。