Keras.backend.reshape:TypeError:无法转换< class'list'>类型的对象.到张量.考虑将元素强制转换为受支持的类型 [英] Keras.backend.reshape: TypeError: Failed to convert object of type <class 'list'> to Tensor. Consider casting elements to a supported type
本文介绍了Keras.backend.reshape:TypeError:无法转换< class'list'>类型的对象.到张量.考虑将元素强制转换为受支持的类型的处理方法，对大家解决问题具有一定的参考价值，需要的朋友们下面随着小编来一起学习吧！
问题描述
我正在为我的神经网络设计一个自定义层，但是我从代码中得到了一个错误.
I'm designing a custom layer for my neural network, but I get an error from my code.
我想按照论文中的描述进行关注: SAGAN .而原始tf代码
I want to do a attention layer as described in the paper: SAGAN. And the original tf code
class AttentionLayer(Layer):
def __init__(self, **kwargs):
super(AttentionLayer, self).__init__(**kwargs)
def build(self, input_shape):
input_dim = input_shape[-1]
filters_f_g = input_dim // 8
filters_h = input_dim
kernel_shape_f_g = (1, 1) + (input_dim, filters_f_g)
kernel_shape_h = (1, 1) + (input_dim, filters_h)
# Create a trainable weight variable for this layer:
self.gamma = self.add_weight(name='gamma', shape=[1], initializer='zeros', trainable=True)
self.kernel_f = self.add_weight(shape=kernel_shape_f_g,
initializer='glorot_uniform',
name='kernel')
self.kernel_g = self.add_weight(shape=kernel_shape_f_g,
initializer='glorot_uniform',
name='kernel')
self.kernel_h = self.add_weight(shape=kernel_shape_h,
initializer='glorot_uniform',
name='kernel')
self.bias_f = self.add_weight(shape=(filters_f_g,),
initializer='zeros',
name='bias')
self.bias_g = self.add_weight(shape=(filters_f_g,),
initializer='zeros',
name='bias')
self.bias_h = self.add_weight(shape=(filters_h,),
initializer='zeros',
name='bias')
super(AttentionLayer, self).build(input_shape)
def call(self, x):
def hw_flatten(x):
return K.reshape(x, shape=[x.shape[0], x.shape[1]*x.shape[2], x.shape[-1]])
f = K.conv2d(x, kernel=self.kernel_f, strides=(1, 1), padding='same') # [bs, h, w, c']
f = K.bias_add(f, self.bias_f)
g = K.conv2d(x, kernel=self.kernel_g, strides=(1, 1), padding='same') # [bs, h, w, c']
g = K.bias_add(g, self.bias_g)
h = K.conv2d(x, kernel=self.kernel_h, strides=(1, 1), padding='same') # [bs, h, w, c]
h = K.bias_add(h, self.bias_h)
# N = h * w
flatten_g = hw_flatten(g)
flatten_f = hw_flatten(f)
s = K.batch_dot(flatten_g, flatten_f, axes=1) # # [bs, N, N]
beta = K.softmax(s, axis=-1) # attention map
o = K.batch_dot(beta, hw_flatten(h)) # [bs, N, C]
o = K.reshape(o, shape=x.shape) # [bs, h, w, C]
x = self.gamma * o + x
return x
当我在模型中添加此层时，出现错误:
When I added this layer in my model, I get an error:
TypeError: Expected binary or unicode string, got Dimension(None)
During handling of the above exception, another exception occurred:
TypeError Traceback (most recent call last)
<ipython-input-5-9d4e83945ade> in <module>()
5 X = Conv2D(64, kernel_size=5, strides=1, name='conv1')(X)
6 X = Activation('relu')(X)
----> 7 X = AttentionLayer()(X)
8 X = Flatten(name='flatten2')(X)
9 X = Dense(1000, activation='relu')(X)
/anaconda3/envs/pycharm/lib/python3.6/site-packages/keras/engine/topology.py in __call__(self, inputs, **kwargs)
617
618 # Actually call the layer, collecting output(s), mask(s), and shape(s).
--> 619 output = self.call(inputs, **kwargs)
620 output_mask = self.compute_mask(inputs, previous_mask)
621
~/Projects/inpainting/models/attention.py in call(self, x)
49
50 # N = h * w
---> 51 flatten_g = hw_flatten(g)
52 flatten_f = hw_flatten(f)
53 s = K.batch_dot(flatten_g, flatten_f, axes=1) # # [bs, N, N]
~/Projects/inpainting/models/attention.py in hw_flatten(x)
39 def call(self, x):
40 def hw_flatten(x):
---> 41 return K.reshape(x, shape=[x.shape[0], x.shape[1]*x.shape[2], x.shape[-1]])
42
43 f = K.conv2d(x, kernel=self.kernel_f, strides=(1, 1), padding='same') # [bs, h, w, c']
/anaconda3/envs/pycharm/lib/python3.6/site-packages/keras/backend/tensorflow_backend.py in reshape(x, shape)
1896 A tensor.
1897 """
-> 1898 return tf.reshape(x, shape)
1899
1900
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/ops/gen_array_ops.py in reshape(tensor, shape, name)
6111 if _ctx is None or not _ctx._eager_context.is_eager:
6112 _, _, _op = _op_def_lib._apply_op_helper(
-> 6113 "Reshape", tensor=tensor, shape=shape, name=name)
6114 _result = _op.outputs[:]
6115 _inputs_flat = _op.inputs
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
511 except TypeError as err:
512 if dtype is None:
--> 513 raise err
514 else:
515 raise TypeError(
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/op_def_library.py in _apply_op_helper(self, op_type_name, name, **keywords)
508 dtype=dtype,
509 as_ref=input_arg.is_ref,
--> 510 preferred_dtype=default_dtype)
511 except TypeError as err:
512 if dtype is None:
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/ops.py in internal_convert_to_tensor(value, dtype, name, as_ref, preferred_dtype, ctx)
1102
1103 if ret is None:
-> 1104 ret = conversion_func(value, dtype=dtype, name=name, as_ref=as_ref)
1105
1106 if ret is NotImplemented:
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py in _constant_tensor_conversion_function(v, dtype, name, as_ref)
233 as_ref=False):
234 _ = as_ref
--> 235 return constant(v, dtype=dtype, name=name)
236
237
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/constant_op.py in constant(value, dtype, shape, name, verify_shape)
212 tensor_value.tensor.CopyFrom(
213 tensor_util.make_tensor_proto(
--> 214 value, dtype=dtype, shape=shape, verify_shape=verify_shape))
215 dtype_value = attr_value_pb2.AttrValue(type=tensor_value.tensor.dtype)
216 const_tensor = g.create_op(
/anaconda3/envs/pycharm/lib/python3.6/site-packages/tensorflow/python/framework/tensor_util.py in make_tensor_proto(values, dtype, shape, verify_shape)
519 raise TypeError("Failed to convert object of type %s to Tensor. "
520 "Contents: %s. Consider casting elements to a "
--> 521 "supported type." % (type(values), values))
522 tensor_proto.string_val.extend(str_values)
523 return tensor_proto
TypeError: Failed to convert object of type <class 'list'> to Tensor. Contents: [Dimension(None), Dimension(64), Dimension(8)]. Consider casting elements to a supported type.
我试图在hw_flatten函数中创建x_shape = x.shape.as_list()，但是它不起作用，我不知道如何调试此错误.
I tried to make x_shape = x.shape.as_list() in the hw_flatten function, but it doesn't works, I have no idea how to debug this error.
推荐答案
您正在访问张量的.shape属性，该属性为您提供Dimension对象，而不是实际的形状值.您有2个选择:
You are accessing the tensor's .shape property which gives you Dimension objects and not actually the shape values. You have 2 options:
- 如果您知道形状并且在创建图层时已将其固定，则可以使用
K.int_shape(x)[0]，该值将为整数.但是，如果在创建时形状未知，它将返回None；否则，将返回None.例如，如果batch_size不明. - 如果将在运行时确定形状，则可以使用
K.shape(x)[0]，它将返回一个符号张量，该张量将在运行时保存形状值.
- If you know the shape and it's fixed at layer creation time you can use
K.int_shape(x)[0]which will give the value as an integer. It will however returnNoneif the shape is unknown at creation time; for example if the batch_size is unknown. - If shape will be determined at runtime then you can use
K.shape(x)[0]which will return a symbolic tensor that will hold the shape value at runtime.
这篇关于Keras.backend.reshape:TypeError:无法转换< class'list'>类型的对象.到张量.考虑将元素强制转换为受支持的类型的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
查看全文
