时间分布LSTM结果的前k个分类准确性 [英] Top k categorical accuracy for Time Distributed LSTM results
问题描述
我正在尝试使用top_k_categorical_accuracy
评估LSTM的结果.
I'm trying to evaluate the results of an LSTM using top_k_categorical_accuracy
.
对于每个一键编码"令牌,我尝试预测下一个令牌.为此,我使用TimeDistributed
层包装器获取序列中每个实例的输出,并将其传递到Dense层以将结果重新编码为相同的一键编码.
For each One-Hot encoded token, I try to predict the next token. In order to do this I take the output for each instance in the sequence by using the TimeDistributed
layer wrapper, and pass it to a Dense layer to re-encode the results into the same One-Hot encoding.
虽然使用内置精度度量标准metrics=['accuracy']
可以顺利进行,但是使用top_k_categorical_accuracy
失败,出现错误消息:
While using the built in accuracy metric metrics=['accuracy']
works without a hitch, using top_k_categorical_accuracy
fails, giving me the error message:
ValueError: Shape must be rank 2 but is rank 3 for 'metrics/my_acc/in_top_k/InTopKV2' (op: 'InTopKV2') with input shapes: [?,?,501], [?,?], [].
要使此指标有效,我需要更改什么?
What do I need to change in order to make this metric work?
我的代码如下:
import numpy as np
import glob
import keras
from keras.models import Sequential
from keras.layers import LSTM, Dense, TimeDistributed,Lambda, Dropout, Activation
from keras.metrics import top_k_categorical_accuracy
train_val_split=0.2 # portion to be placed in validation
train_control_number=0
val_control_number=0
def my_acc(y_true, y_pred):
return top_k_categorical_accuracy(y_true, y_pred, k=5)
def basic_LSTM(features_num):
model = Sequential()
model.add(LSTM(40, return_sequences=True, input_shape=(None, features_num)))
model.add(LSTM(40, return_sequences=True))
model.add(LSTM(40, return_sequences=True))
model.add(TimeDistributed(Dense(features_num)))
model.add(Activation('linear'))
print(model.summary())
model.compile(loss='categorical_crossentropy', optimizer='adam',metrics=[my_acc])
return (model)
def main ():
input_files=glob.glob('*npy')
data_list,dim=loader(input_files)
train_list,val_list=data_spliter(data_list)
model=basic_LSTM(dim)
model.fit_generator(train_generator(train_list), steps_per_epoch=len(train_list), epochs=10, verbose=1,validation_data=val_generator(val_list),validation_steps=len(val_list))
def train_generator(data_list):
while True:
global train_control_number
train_control_number=cycle_throught(len(data_list),train_control_number)
this=data_list[train_control_number]
x_train = this [:,:-1,:] # all but the last 1
y_train = this [:,1:,:] # all but the first 1
yield (x_train, y_train)
def val_generator(data_list):
while True:
global val_control_number
val_control_number=cycle_throught(len(data_list),val_control_number)
this=data_list[val_control_number]
x_train = this [:,:-1,:] # all but the last 1
y_train = this [:,1:,:] # all but the first 1
yield (x_train, y_train)
def cycle_throught (total,current):
current+=1
if (current==total):
current=0
return (current)
def loader(input_files):
data_list=[]
for input_file in input_files:
a=np.load (input_file)
incoming_shape=list(a.shape)
requested_shape=[1]+incoming_shape
a=a.reshape(requested_shape)
data_list.append(a)
return (data_list,incoming_shape[-1])
def data_spliter(input_list):
val_num=int(len(input_list)*train_val_split)
validation=input_list[:val_num]
train=input_list[val_num:]
return (train,validation)
main()
非常感谢.
推荐答案
您可以在自定义指标内以2D张量变换数据,使其适合所需的形状,而使最后一个轴保持不变:
You can transform your data in a 2D tensor within a custom metric so it suits to the required shape, keeping the last axis untouched:
import keras.backend as K #or tf.keras.backend as K
def 3D_top_k(true, pred):
true = K.reshape(true, (-1, features_num))
pred = K.reshape(pred, (-1, features_num))
return top_k_categorical_accuracy(true, pred, k=5)
这篇关于时间分布LSTM结果的前k个分类准确性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!