ValueError: Argument must be a dense tensor:... got shape [6, 60, 160, 3], but wanted [6].

在将 列表或元组 数据转换成 dataset类型时

import numpy as np
import tensorflow as tf
from sklearn.cross_validation import train_test_split

pic_array=np.ones((60,160,3)) #图片的长宽为60*160，每个像素点的由rgb3个值表示像素
pic_txt_array=np.ones((26,4)) #表示单个字母的向量长为26，共4个字母
data_x=[pic_array  for i in range(1000)] #1000张图片的集合
data_y=[pic_txt_array for i in range(1000)]#1000长图片对应的字母的集合

#将装着样本的列表 转换成dataset格式
train_dataset=tf.data.Dataset.from_tensor_slices((data_x,data_y))

发生异常：

File "tf_test.py", line 10, in <module>
train_dataset=tf.data.Dataset.from_tensor_slices((data_x,data_y))
File "F:Program FilesPython35libsite-packages ensorflowpythondataopsdataset_ops.py", line 235, in from_tensor_slices
return TensorSliceDataset(tensors)
File "F:Program FilesPython35libsite-packages ensorflowpythondataopsdataset_ops.py", line 1030, in __init__
for i, t in enumerate(nest.flatten(tensors))
........
File "F:Program FilesPython35libsite-packages ensorflowpythonframeworkconstant_op.py", line 214, in constant
value, dtype=dtype, shape=shape, verify_shape=verify_shape))
File "F:Program FilesPython35libsite-packages ensorflowpythonframework ensor_util.py", line 441, in make_tensor_proto
_GetDenseDimensions(values)))
ValueError: Argument must be a dense tensor: [array([[[1., 1., 1.],...

...,
[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.]]])] - got shape [6, 60, 160, 3], but wanted [6].

解决：修改源数据的格式

data_x=np.asarray([pic_array for i in range(1000)]) #1000张图片的集合
data_y=np.asarray([pic_txt_array for i in range(1000)]) #1000长图片对应的字母的集合