我有一个NN有两个相同的CNN(类似于Siamese网络),然后合并输出,并打算在合并输出上应用自定义丢失函数,如下所示:
----------------- -----------------
| input_a | | input_b |
----------------- -----------------
| base_network | | base_network |
------------------------------------------
| processed_a_b |
------------------------------------------
在我的自定义丢失函数中,我需要将y垂直分成两部分,然后在每个部分上应用分类交叉熵损失。但是,我不断从我的损失函数中得到dtype错误,例如:
ValueError Traceback(最近一次调用 最后)in() ----> 1个model.compile(loss = categorical_crossentropy_loss,optimizer = RMSprop())
/usr/local/lib/python3.5/dist-packages/keras/engine/training.py in 编译(自我,优化器,损失,指标,loss_weights, sample_weight_mode,** kwargs) 909 loss_weight = loss_weights_list [i] 910 output_loss = weighted_loss(y_true,y_pred, - > 911 sample_weight,mask) 912如果len(self.outputs)> 1: 913 self.metrics_tensors.append(output_loss)
/usr/local/lib/python3.5/dist-packages/keras/engine/training.py in 加权(y_true,y_pred,权重,掩码) 451#应用样本加权 452如果权重不是无: - > 453 score_array * =重量 454 score_array / = K.mean(K.cast(K.not_equal(weights,0),K.floatx())) 455返回K.mean(score_array)
/usr/local/lib/python3.5/dist-packages/tensorflow/python/ops/math_ops.py 在binary_op_wrapper(x,y)中 827如果不是isinstance(y,sparse_tensor.SparseTensor): 828尝试: - > 829 y = ops.convert_to_tensor(y,dtype = x.dtype.base_dtype,name =" y") 830除TypeError外: 831#如果RHS不是张量,它可能是张量识别对象
/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py in convert_to_tensor(value,dtype,name,preferred_dtype) 674 name = name, 675 preferred_dtype = preferred_dtype, - > 676 as_ref = False) 677 678
/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py 在internal_convert_to_tensor中(value,dtype,name,as_ref, preferred_dtype) 739 740如果ret为None: - > 741 ret = conversion_func(value,dtype = dtype,name = name,as_ref = as_ref) 742 743如果ret未实现:
/usr/local/lib/python3.5/dist-packages/tensorflow/python/framework/ops.py 在_TensorTensorConversionFunction(t,dtype,name,as_ref)中 612提出ValueError( 613"张量转换为Tensor请求dtype%s,dtype为%s:%r" - > 614%(dtype.name,t.dtype.name,str(t))) 615返回t 616
ValueError:Tensor转换为Tensor请求dtype float64 dtype float32:' Tensor(" processed_a_b_sample_weights_1:0",shape =(?,), D型= FLOAT32)'
这是一个重现错误的MWE:
import tensorflow as tf
from keras import backend as K
from keras.layers import Input, Dense, merge, Dropout
from keras.models import Model, Sequential
from keras.optimizers import RMSprop
import numpy as np
# define the inputs
input_dim = 10
input_a = Input(shape=(input_dim,), name='input_a')
input_b = Input(shape=(input_dim,), name='input_b')
# define base_network
n_class = 4
base_network = Sequential(name='base_network')
base_network.add(Dense(8, input_shape=(input_dim,), activation='relu'))
base_network.add(Dropout(0.1))
base_network.add(Dense(n_class, activation='relu'))
processed_a = base_network(input_a)
processed_b = base_network(input_b)
# merge left and right sections
processed_a_b = merge([processed_a, processed_b], mode='concat', concat_axis=1, name='processed_a_b')
# create the model
model = Model(inputs=[input_a, input_b], outputs=processed_a_b)
# custom loss function
def categorical_crossentropy_loss(y_true, y_pred):
# break (un-merge) y_true and y_pred into two pieces
y_true_a, y_true_b = tf.split(value=y_true, num_or_size_splits=2, axis=1)
y_pred_a, y_pred_b = tf.split(value=y_pred, num_or_size_splits=2, axis=1)
loss = K.categorical_crossentropy(output=y_pred_a, target=y_true_a) + K.categorical_crossentropy(output=y_pred_b, target=y_true_b)
return K.mean(loss)
# compile the model
model.compile(loss=categorical_crossentropy_loss, optimizer=RMSprop())
答案 0 :(得分:0)
如您的错误所示,您正在处理float32
数据,并且预计会float64
。有必要将错误跟踪到其特定行,以确定要纠正的张量,并能够更好地帮助您。
但是,似乎与K.mean()
方法相关,但ValueError
方法也可以生成K.categorical_crossentropy()
。因此问题可能在于您的张量loss
,y_pred
或y_true
。鉴于这些情况,我看到你可以尝试解决问题的两件事: