- html - 出于某种原因,IE8 对我的 Sass 文件中继承的 html5 CSS 不友好?
- JMeter 在响应断言中使用 span 标签的问题
- html - 在 :hover and :active? 上具有不同效果的 CSS 动画
- html - 相对于居中的 html 内容固定的 CSS 重复背景?
我正在研究序列预测问题,但我在这方面没有太多经验,所以下面的一些问题可能很幼稚。
仅供引用:我创建了一个关注 CRF 的后续问题 here
我有以下问题:
我想预测多个非独立变量的二进制序列。
输入:
我有一个包含以下变量的数据集:
binary_signal_group_A
和
binary_signal_group_B
是我想使用(1)它们过去的行为和(2)从每个时间戳中提取的附加信息来预测的 2 个非独立变量。
# required libraries
import re
import numpy as np
import pandas as pd
from keras import Sequential
from keras.layers import LSTM
data_length = 18 # how long our data series will be
shift_length = 3 # how long of a sequence do we want
df = (pd.DataFrame # create a sample dataframe
.from_records(np.random.randint(2, size=[data_length, 3]))
.rename(columns={0:'a', 1:'b', 2:'extra'}))
# NOTE: the 'extra' variable refers to a generic predictor such as for example 'is_weekend' indicator, it doesn't really matter what it is
# shift so that our sequences are in rows (assuming data is sorted already)
colrange = df.columns
shift_range = [_ for _ in range(-shift_length, shift_length+1) if _ != 0]
for c in colrange:
for s in shift_range:
if not (c == 'extra' and s > 0):
charge = 'next' if s > 0 else 'last' # 'next' variables is what we want to predict
formatted_s = '{0:02d}'.format(abs(s))
new_var = '{var}_{charge}_{n}'.format(var=c, charge=charge, n=formatted_s)
df[new_var] = df[c].shift(s)
# drop unnecessary variables and trim missings generated by the shift operation
df.dropna(axis=0, inplace=True)
df.drop(colrange, axis=1, inplace=True)
df = df.astype(int)
df.head() # check it out
# a_last_03 a_last_02 ... extra_last_02 extra_last_01
# 3 0 1 ... 0 1
# 4 1 0 ... 0 0
# 5 0 1 ... 1 0
# 6 0 0 ... 0 1
# 7 0 0 ... 1 0
# [5 rows x 15 columns]
# separate predictors and response
response_df_dict = {}
for g in ['a','b']:
response_df_dict[g] = df[[c for c in df.columns if 'next' in c and g in c]]
# reformat for LSTM
# the response for every row is a matrix with depth of 2 (the number of groups) and width = shift_length
# the predictors are of the same dimensions except the depth is not 2 but the number of predictors that we have
response_array_list = []
col_prefix = set([re.sub('_\d+$','',c) for c in df.columns if 'next' not in c])
for c in col_prefix:
current_array = df[[z for z in df.columns if z.startswith(c)]].values
response_array_list.append(current_array)
# reshape into samples (1), time stamps (2) and channels/variables (0)
response_array = np.array([response_df_dict['a'].values,response_df_dict['b'].values])
response_array = np.reshape(response_array, (response_array.shape[1], response_array.shape[2], response_array.shape[0]))
predictor_array = np.array(response_array_list)
predictor_array = np.reshape(predictor_array, (predictor_array.shape[1], predictor_array.shape[2], predictor_array.shape[0]))
# feed into the model
model = Sequential()
model.add(LSTM(8, input_shape=(predictor_array.shape[1],predictor_array.shape[2]), return_sequences=True)) # the number of neurons here can be anything
model.add(LSTM(2, return_sequences=True)) # should I use an activation function here? the number of neurons here must be equal to the # of groups we are predicting
model.summary()
# _________________________________________________________________
# Layer (type) Output Shape Param #
# =================================================================
# lstm_62 (LSTM) (None, 3, 8) 384
# _________________________________________________________________
# lstm_63 (LSTM) (None, 3, 2) 88
# =================================================================
# Total params: 472
# Trainable params: 472
# Non-trainable params: 0
model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy']) # is it valid to use crossentropy and accuracy as metric?
model.fit(predictor_array, response_array, epochs=10, batch_size=1)
model_preds = model.predict_classes(predictor_array) # not gonna worry about train/test split here
model_preds.shape # should return (12, 3, 2) or (# of records, # of timestamps, # of groups which are a and b)
# (12, 3)
model_preds
# array([[1, 0, 0],
# [0, 0, 0],
# [1, 0, 0],
# [0, 0, 0],
# [1, 0, 0],
# [0, 0, 0],
# [0, 0, 0],
# [0, 0, 0],
# [0, 0, 0],
# [0, 0, 0],
# [1, 0, 0],
# [0, 0, 0]])
最佳答案
我会依次回答所有问题
how do I get this working so that the model would forecast the next N sequences for both groups?
L
是计算损失,
p
是网络预测和
y
是目标值。
nan
为损失。
model = Sequential()
model.add(LSTM(8, input_shape=(predictor_array.shape[1],predictor_array.shape[2]), return_sequences=True))
model.add(LSTM(2, return_sequences=True))
model.add(TimeDistributed(Dense(2, activation="sigmoid")))
model.summary()
... is it valid to attempt to output both A and B sequences by a single model or should I fit 2 separate models ... ?
... the prediction output is of shape (12, 3) when I would have expected it to be (12, 2) -- am I doing something wrong here ... ?
As far as the output LSTM layer is concerned, would it be a good idea to use an activation function here, such as sigmoid? Why/why not?
Is it valid to use a classification type loss (binary cross-entropy) and metrics (accuracy) for optimizing a sequence?
Is an LSTM model an optimal choice here? Does anyone think that a CRF or some HMM-type model would work better here?
关于tensorflow - 使用 LSTM 进行多元二进制序列预测,我们在Stack Overflow上找到一个类似的问题: https://stackoverflow.com/questions/53977695/
我喜欢 smartcase,也喜欢 * 和 # 搜索命令。但我更希望 * 和 # 搜索命令区分大小写,而/和 ?搜索命令遵循 smartcase 启发式。 是否有隐藏在某个地方我还没有找到的设置?我宁
关闭。这个问题是off-topic .它目前不接受答案。 想改进这个问题? Update the question所以它是on-topic对于堆栈溢出。 10年前关闭。 Improve this qu
从以下网站,我找到了执行java AD身份验证的代码。 http://java2db.com/jndi-ldap-programming/solution-to-sslhandshakeexcepti
似乎 melt 会使用 id 列和堆叠的测量变量 reshape 您的数据框,然后通过转换让您执行聚合。 ddply,从 plyr 包看起来非常相似..你给它一个数据框,几个用于分组的列变量和一个聚合
我的问题是关于 memcached。 Facebook 使用 memcached 作为其结构化数据的缓存,以减少用户的延迟。他们在 Linux 上使用 UDP 优化了 memcached 的性能。 h
在 Camel route ,我正在使用 exec 组件通过 grep 进行 curl ,但使用 ${HOSTNAME} 的 grep 无法正常工作,下面是我的 Camel 路线。请在这方面寻求帮助。
我正在尝试执行相当复杂的查询,在其中我可以排除与特定条件集匹配的项目。这是一个 super 简化的模型来解释我的困境: class Thing(models.Model) user = mod
我正在尝试执行相当复杂的查询,我可以在其中排除符合特定条件集的项目。这里有一个 super 简化的模型来解释我的困境: class Thing(models.Model) user = mod
我发现了很多嵌入/内容项目的旧方法,并且我遵循了在这里找到的最新方法(我假设):https://blog.angular-university.io/angular-ng-content/ 我正在尝试
我正在寻找如何使用 fastify-nextjs 启动 fastify-cli 的建议 我曾尝试将代码简单地添加到建议的位置,但它不起作用。 'use strict' const path = req
我正在尝试将振幅 js 与 React 和 Gatsby 集成。做 gatsby developer 时一切看起来都不错,因为它发生在浏览器中,但是当我尝试 gatsby build 时,我收到以下错
我试图避免过度执行空值检查,但同时我想在需要使代码健壮的时候进行空值检查。但有时我觉得它开始变得如此防御,因为我没有实现 API。然后我避免了一些空检查,但是当我开始单元测试时,它开始总是等待运行时异
尝试进行包含一些 NOT 的 Kibana 搜索,但获得包含 NOT 的结果,因此猜测我的语法不正确: "chocolate" AND "milk" AND NOT "cow" AND NOT "tr
我正在使用开源代码共享包在 iOS 中进行 facebook 集成,但收到错误“FT_Load_Glyph failed: glyph 65535: error 6”。我在另一台 mac 机器上尝试了
我正在尝试估计一个标准的 tobit 模型,该模型被审查为零。 变量是 因变量 : 幸福 自变量 : 城市(芝加哥,纽约), 性别(男,女), 就业(0=失业,1=就业), 工作类型(失业,蓝色,白色
我有一个像这样的项目布局 样本/ 一种/ 源/ 主要的/ java / java 资源/ .jpg 乙/ 源/ 主要的/ java / B.java 资源/ B.jpg 构建.gradle 设置.gr
如何循环遍历数组中的多个属性以及如何使用map函数将数组中的多个属性显示到网页 import React, { Component } from 'react'; import './App.css'
我有一个 JavaScript 函数,它进行 AJAX 调用以返回一些数据,该调用是在选择列表更改事件上触发的。 我尝试了多种方法来在等待时显示加载程序,因为它当前暂停了选择列表,从客户的 Angul
可能以前问过,但找不到。 我正在用以下形式写很多语句: if (bar.getFoo() != null) { this.foo = bar.getFoo(); } 我想到了三元运算符,但我认
我有一个表单,在将其发送到 PHP 之前我正在执行一些验证 JavaScript,验证后的 JavaScript 函数会发布用户在 中输入的文本。页面底部的标签;然而,此消息显示短暂,然后消失...
我是一名优秀的程序员,十分优秀!