markilue
/
self_example
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

leecode更新

This commit is contained in:
markilue 2022-10-28 17:00:14 +08:00
parent 4b279d5f1d
commit 3a2eadc687
6 changed files with 1473 additions and 98 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

125
Leecode/src/main/java/com/markilue/leecode/greedy/Jump.java Normal file
View File

@ -0,0 +1,125 @@
package com.markilue.leecode.greedy;
import org.junit.Test;
import java.util.ArrayList;
import java.util.HashSet;
/**
* @BelongsProject: Leecode
* @BelongsPackage: com.markilue.leecode.greedy
* @Author: markilue
* @CreateTime: 2022-10-28 10:15
* @Description: TODO 力扣45题 跳跃游戏II：
* 给你一个非负整数数组nums ，你最初位于数组的第一个位置。
* 数组中的每个元素代表你在该位置可以跳跃的最大长度。
* 你的目标是使用最少的跳跃次数到达数组的最后一个位置。
* @Version: 1.0
*/
public class Jump {
@Test
public void test() {
int[] nums = {2, 2, 2, 1, 4};
System.out.println(jump(nums));
}
@Test
public void test1() {
int[] nums = {2, 3, 1, 1, 4};
System.out.println(jump(nums));
}
@Test
public void test2() {
int[] nums = {1, 1, 1, 1, 1};
System.out.println(jump(nums));
}
@Test
public void test3() {
int[] nums = {1, 1};
System.out.println(jump(nums));
}
@Test
public void test4() {
int[] nums = {3, 2, 1};
System.out.println(jump(nums));
}
@Test
public void test5() {
int[] nums = {5, 4, 0, 1, 3, 6, 8, 0, 9, 4, 9, 1, 8, 7, 4, 8};
System.out.println(jump1(nums));
}
/**
* 要求使用最少跳跃次数达到数组的最后一个位置：即最早能覆盖cover到最后
* 速度击败99.04%，内存击败62.81%
*
* @param nums
* @return
*/
public int jump(int[] nums) {
if (nums.length == 1) {
return 0;
}
int curDistance = 0; //当前覆盖的最远距离的下标
int step = 0; //记录要走的步数
int nextDistance = 0; //下一步覆盖的最远距离的下标
for (int i = 0; i < nums.length; i++) {
//更新下一步覆盖的最远距离的下标
nextDistance = Math.max(nums[i] + i, nextDistance);
//遇到当前覆盖的最远距离的下标
if (i == curDistance) {
if (curDistance != nums.length - 1) {
step++;//需要走一步
curDistance = nextDistance;//更新当前覆盖的最远距离的下标
//下一步的覆盖范围已经包含终点，结束循环
if (nextDistance >= nums.length - 1) break;
} else break;//当前覆盖的最远距离的下标是集合终点，不需要再走了
}
}
return step;
}
/**
* 贪心算法II:即移动下标只要遇到当前覆盖的最远距离的下标，则步数加一，不考虑是不是终点的情况
* 速度击败99.04%，内存击败62.81%
*
* @param nums
* @return
*/
public int jump1(int[] nums) {
int curDistance = 0; //当前覆盖的最远距离的下标
int step = 0; //记录要走的步数
int nextDistance = 0; //下一步覆盖的最远距离的下标
//注意这里是小于nums.size()-1，这里是关键
for (int i = 0; i < nums.length - 1; i++) {
//更新下一步覆盖的最远距离的下标
nextDistance = Math.max(nums[i] + i, nextDistance);
//遇到当前覆盖的最远距离的下标
if (i == curDistance) {
curDistance = nextDistance;
step++;//需要走一步
}
}
return step;
}
}

267
TensorFlow_eaxmple/Model_train_test/condition_monitoring/plot/test_plot.py
View File

@ -8,6 +8,13 @@ import random
import pandas as pd
import seaborn as sns
from condition_monitoring.data_deal.loadData import read_data
from model.Joint_Monitoring.Joint_Monitoring_banda import Joint_Monitoring
from model.Joint_Monitoring.compare.RNet_L import Joint_Monitoring as Joint_Monitoring_L
from model.Joint_Monitoring.compare.RNet_S import Joint_Monitoring as Joint_Monitoring_SE
import tensorflow as tf
import tensorflow.keras
from mpl_toolkits.axes_grid1.inset_locator import mark_inset
from mpl_toolkits.axes_grid1.inset_locator import inset_axes
'''
@Author : dingjiawen
@ -24,7 +31,12 @@ max_file_name = "E:\self_example\TensorFlow_eaxmple\Model_train_test\condition_m
source_path = "G:\data\SCADA数据\jb4q_8_delete_total_zero.csv"
list = [64.16, 65.26, 65.11, 66.6, 67.16, 66.28, 73.86, 75.24, 73.98, 76.7, 98.86, 99.45, 99.97]
model_name = "../hard_model/two_weight/banda_joint_epoch0_8887_9454/weight"
model_name1 = "../hard_model/two_weight/banda_joint_epoch14_9872_9863/weight"
file_name = "E:\跑模型\论文写作/DCAU.txt"
# list = [64.16, 65.26, 65.11, 66.6, 67.16, 66.28, 73.86, 75.24, 73.98, 76.7, 98.86, 99.45, 99.97]
def plot_result(result_data):
@ -39,10 +51,10 @@ def plot_result(result_data):
'legend.fontsize': 5,
}
plt.rcParams.update(parameters)
plt.figure()
plt.rc('font', family='Times New Roman') # 全局字体样式#画混淆矩阵
fig,ax=plt.subplots(1, 1)
plt.rc('font', family='Times New Roman') # 全局字体样式
font1 = {'family': 'Times New Roman', 'weight': 'normal', 'size': 5} # 设置坐标标签的字体大小，字体
plt.scatter(list(range(result_data.shape[0])), result_data, c='black', s=0.5, label="predict")
plt.scatter(list(range(result_data.shape[0])), result_data, c='black', s=0.01, label="predict")
# 画出 y=1 这条水平线
plt.axhline(0.5, c='red', label='Failure threshold', lw=1)
# 箭头指向上面的水平线
@ -79,9 +91,49 @@ def plot_result(result_data):
plt.ylabel('Confidence', fontsize=5)
plt.xlabel('Time', fontsize=5)
plt.tight_layout()
# plt.legend(loc='best',edgecolor='black',fontsize=3)
plt.legend(loc='best', frameon=False, fontsize=3)
# plt.legend(loc='best', edgecolor='black', fontsize=4)
plt.legend(loc='best', frameon=False, fontsize=4)
# plt.grid()
#局部方法图
axins = inset_axes(ax, width="40%", height="30%", loc='lower left',
bbox_to_anchor=(0.1, 0.1, 1, 1),
bbox_transform=ax.transAxes)
axins.scatter(list(range(result_data.shape[0])), result_data, c='black', s=0.001, label="predict")
axins.axvline(result_data.shape[0] * 2 / 3, c='blue', ls='-.', lw=0.5, label='real fault')
plt.axhline(0.5, c='red', label='Failure threshold', lw=0.5)
#设置放大区间
# 设置放大区间
zone_left = int(result_data.shape[0]*2/3-100)
zone_right = int(result_data.shape[0]*2/3)+100
x=list(range(result_data.shape[0]))
# 坐标轴的扩展比例（根据实际数据调整）
x_ratio = 0.5 # x轴显示范围的扩展比例
y_ratio = 0.5 # y轴显示范围的扩展比例
mark_inset(ax, axins, loc1=4, loc2=2, fc="none", ec='k', lw=0.5)
# X轴的显示范围
xlim0 = x[zone_left] - (x[zone_right] - x[zone_left]) * x_ratio
xlim1 = x[zone_right] + (x[zone_right] - x[zone_left]) * x_ratio
# axins.tick_params(bottom=False)
axins.xaxis.set_visible(False)
# Y轴的显示范围
# y = np.hstack((y_1[zone_left:zone_right], y_2[zone_left:zone_right], y_3[zone_left:zone_right]))
# ylim0 = np.min(y) - (np.max(y) - np.min(y)) * y_ratio
# ylim1 = np.max(y) + (np.max(y) - np.min(y)) * y_ratio
# 调整子坐标系的显示范围
axins.set_xlim(xlim0, xlim1)
# axins.set_ylim(ylim0, ylim1)
plt.show()
pass
@ -133,31 +185,31 @@ def plot_MSE(total_MSE, total_max):
def plot_Corr(data, size: int = 1):
parameters = {
'figure.dpi': 600,
'figure.figsize': (2.8 * size, 2 * size),
'figure.figsize': (9, 7),
'savefig.dpi': 600,
'xtick.direction': 'inout',
'ytick.direction': 'inout',
'xtick.labelsize': 3 * size,
'ytick.labelsize': 3 * size,
'legend.fontsize': 5 * size,
'xtick.labelsize': 14,
'ytick.labelsize': 14,
'legend.fontsize': 12,
}
plt.rcParams.update(parameters)
plt.figure()
plt.rc('font', family='Times New Roman') # 全局字体样式#画混淆矩阵
font1 = {'family': 'Times New Roman', 'weight': 'normal', 'size': 4 * size} # 设置坐标标签的字体大小，字体
# font1 = {'family': 'Times New Roman', 'weight': 'normal', 'size': 4 * size} # 设置坐标标签的字体大小，字体
print("计算皮尔逊相关系数")
pd_data = pd.DataFrame(data)
person = pd_data.corr()
print(person)
# 画热点图heatmap
cmap = sns.heatmap(person, annot=True, annot_kws={
'fontsize': 2.6 * size
cmap = sns.heatmap(person, annot=True, cmap='Blues',annot_kws={
'fontsize': 11
})
classes = ['Gs', 'Gio', 'Gip', 'Gp', 'Gwt', 'En', 'Gft', 'Grt', 'Gwt', 'Et', 'Rs', 'Ap', 'Ws', 'Dw', 'Ges', 'Gt',
'Vx', 'Vy']
indices = range(len(person))
plt.title("Heatmap of correlation coefficient matrix", size=6 * size, fontdict=font1)
plt.title("Heatmap of correlation coefficient matrix", size=18)
# pad调整label与坐标轴之间的距离
plt.tick_params(bottom=False, top=False, left=False, right=False, direction='inout', length=2, width=0.5, pad=1)
plt.xticks([index + 0.5 for index in indices], classes, rotation=0) # 设置横坐标方向，rotation=45为45度倾斜
@ -165,11 +217,11 @@ def plot_Corr(data, size: int = 1):
# 调整色带的标签：
cbar = cmap.collections[0].colorbar
cbar.ax.tick_params(labelsize=4 * size, labelcolor="black", length=2, width=0.5, pad=1)
cbar.ax.set_ylabel(ylabel="color scale", color="black", loc="center", fontdict=font1)
cbar.ax.tick_params(labelsize=10, labelcolor="black", length=2, width=0.5, pad=1)
# cbar.ax.set_ylabel(ylabel="color scale", color="black", loc="center", size=12)
# plt.axis('off') # 去坐标轴
plt.savefig('./corr.png')
# plt.savefig('./corr.png')
plt.show()
pass
@ -220,6 +272,7 @@ def plot_bar(y_data):
plt.ylim([-0.01, 5])
plt.show()
def acc(y_data=list):
parameters = {
'figure.dpi': 600,
@ -227,9 +280,9 @@ def acc(y_data=list):
'savefig.dpi': 600,
'xtick.direction': 'in',
'ytick.direction': 'in',
'xtick.labelsize': 15,
'ytick.labelsize': 15,
'legend.fontsize': 12,
'xtick.labelsize': 18,
'ytick.labelsize': 20,
'legend.fontsize': 14,
}
plt.rcParams.update(parameters)
plt.figure()
@ -255,15 +308,16 @@ def acc(y_data=list):
# plt.tick_params(bottom=False, top=False, left=True, right=False, direction='in', pad=1)
plt.xticks([])
plt.ylabel('Accuracy(%)',fontsize=18)
# plt.xlabel('Time', fontsize=5)
plt.ylabel('Accuracy(%)', fontsize=20)
plt.xlabel('Methods', fontsize=20)
# plt.tight_layout()
num1, num2, num3, num4 = 0.08, 1, 3, 0
num1, num2, num3, num4 = 0, 1, 3, 0
plt.legend(bbox_to_anchor=(num1, num2), loc=num3, borderaxespad=num4, ncol=5, frameon=False, markerscale=0.5)
plt.ylim([60, 105])
plt.show()
def plot_FNR1(y_data):
parameters = {
'figure.dpi': 600,
@ -348,6 +402,84 @@ def plot_FNR2(y_data):
plt.show()
def plot_hot(data):
# 画热点图heatmap
plt.figure(1, figsize=(9, 12))
for i in range(8):
plt.subplot(8, 1, i + 1)
# cbar_kws设置色带，fraction设置大小，pad设置填充
cmap = sns.heatmap(np.expand_dims(data[i, :], axis=0), annot=False, cmap='Blues', linecolor='black',
linewidths=1,
cbar_kws={"orientation": "horizontal",
'fraction': 0.6,
'pad': 0.00
})
plt.title("Pred=0,Expc=0", size=18)
plt.axis('off') # 去坐标轴
# 调整色带的标签：
cbar = cmap.collections[0].colorbar
cbar.ax.tick_params(labelsize=15, labelcolor="black")
# cbar.ax.set_xlabel(xlabel="color scale", color="red", loc="center", fontdict=font2)
# plt.tight_layout()
plt.show()
pass
def plot_hot_one(data):
# 画热点图heatmap
# plt.figure(1, figsize=(9, 12))
plt.subplots(figsize=(14,2))
# cbar_kws设置色带，fraction设置大小，pad设置填充
cmap = sns.heatmap(data, annot=False, cmap='Blues', linecolor='black', linewidths=1,
cbar_kws={"orientation": "horizontal",
'fraction': 0.6,
'pad': 0.00
})
plt.title("Pred=0,Expc=0", size=18)
plt.axis('off') # 去坐标轴
# 调整色带的标签：
cbar = cmap.collections[0].colorbar
cbar.ax.tick_params(labelsize=15, labelcolor="black")
# cbar.ax.set_xlabel(xlabel="color scale", color="red", loc="center", fontdict=font2)
# plt.tight_layout()
plt.show()
pass
def plot_mse(file_name="../others_idea/mse"):
mse=np.loadtxt(file_name,delimiter=",")
mse= mse[2000:2338]
parameters = {
'figure.dpi': 600,
'figure.figsize': (7,5),
'savefig.dpi': 600,
'xtick.direction': 'in',
'ytick.direction': 'in',
'xtick.labelsize': 15,
'ytick.labelsize': 15,
'legend.fontsize': 10,
}
plt.rcParams.update(parameters)
plt.figure()
plt.rc('font', family='Times New Roman') # 全局字体样式#画混淆矩阵
indices = [mse.shape[0] * i / 4 for i in range(5)]
classes = ['13/09/17', '14/09/17', '15/09/17', '16/09/17', '17/09/17']
# plt.xticks([index + 0.5 for index in indices], classes, rotation=25) # 设置横坐标方向，rotation=45为45度倾斜
plt.ylabel('MSE', fontsize=15)
plt.xlabel('Time', fontsize=15)
plt.tight_layout()
plt.plot(mse)
plt.show()
def test_result(file_name: str = result_file_name):
# result_data = np.recfromcsv(file_name)
result_data = np.loadtxt(file_name, delimiter=",")
@ -365,8 +497,8 @@ def test_result(file_name: str = result_file_name):
print("漏报率", negative_rate)
# 画图
data = np.zeros([208, ])
result_data = np.concatenate([result_data, data], axis=0)
# data = np.zeros([208, ])
# result_data = np.concatenate([result_data, data], axis=0)
print(result_data)
print(result_data.shape)
plot_result(result_data)
@ -382,12 +514,15 @@ def test_mse(mse_file_name: str = mse_file_name, max_file_name: str = max_file_n
plot_MSE(mse_data, max_data)
def test_corr(file_name=source_path, N=10):
needed_data, label = read_data(file_name=file_name, isNew=False)
print(needed_data)
print(needed_data.shape)
# needed_data, label = read_data(file_name=file_name, isNew=False)
# print(needed_data)
# print(needed_data.shape)
# np.save("corr.npy",needed_data)
needed_data=np.load("corr.npy")
# plot_original_data(needed_data)
person = plot_Corr(needed_data, size=3)
person = plot_Corr(needed_data)
person = np.array(person)
pass
@ -396,6 +531,66 @@ def test_bar(y_data=list):
plot_bar(y_data)
def test_model_weight(model_name=model_name):
print("===============第一次===================")
model = Joint_Monitoring()
model.load_weights(model_name)
# model.build(input_shape=(32, 120, 10))
DCAU = model.get_layer('dynamic_channel_attention')
DCAU.build(input_shape=[16, 120, 20])
one = tf.ones(shape=[16, 120, 20])
weight = DCAU.call(one)[0, 0, :10]
print(weight)
weight = np.expand_dims(weight, axis=0)
plot_hot(weight)
pass
def test_model_weight_l(model_name=model_name):
print("===============第一次===================")
model = Joint_Monitoring_L()
model.load_weights(model_name)
# model.build(input_shape=(32, 120, 10))
# print(model.summary())
DCAU = model.get_layer('light_channel_attention')
DCAU.build(input_shape=[16, 120, 20])
one = tf.ones(shape=[16, 120, 20])
weight = DCAU.call(one)[0, 0, :10]
print(weight)
weight = np.expand_dims(weight, axis=0)
plot_hot_one(weight)
pass
def test_model_weight_s(model_name=model_name):
print("===============第一次===================")
model = Joint_Monitoring_SE()
model.load_weights(model_name)
# model.build(input_shape=(32, 120, 10))
# print(model.summary())
DCAU = model.get_layer('se_channel_attention')
DCAU.build(input_shape=[16, 120, 20])
one = tf.ones(shape=[16, 120, 20])
weight = DCAU.call(one)[0, 0, :10]
print(weight)
weight = np.expand_dims(weight, axis=0)
plot_hot_one(weight)
pass
def test_model_visualization(model_name=file_name):
with open(file_name, 'r', encoding='utf-8') as f:
data = np.loadtxt(f, str, delimiter=',')
needed_data = data[1:, 1:].astype(dtype=np.float)
print(needed_data.shape)
plot_hot(needed_data)
if __name__ == '__main__':
# test_mse()
# test_result(file_name='E:\self_example\TensorFlow_eaxmple\Model_train_test\condition_monitoring\self_try\compare\mse\RNet_C\RNet_C_timestamp120_feature10_result2.csv')
@ -407,5 +602,15 @@ if __name__ == '__main__':
# list=[99.99,98.95,99.95,96.1,95,99.65,76.25,72.64,75.87,68.74]
# plot_FNR1(list)
#
list=[3.43,1.99,1.92,2.17,1.63,1.81,1.78,1.8,0.6]
plot_FNR2(list)
# list=[3.43,1.99,1.92,2.17,1.63,1.81,1.78,1.8,0.6]
# plot_FNR2(list)
# 查看网络某一层的权重
# test_model_visualization()
file_name = "../self_try\compare\model\weight\RNet_L_banda_epoch17_0.0086_0.0092/weight"
test_model_weight_l(file_name)
# file_name = "../self_try\compare\model\weight\RNet_S_epoch3_2.47_1.63/weight"
# test_model_weight_l(file_name)
#单独预测图
# plot_mse()

579
TensorFlow_eaxmple/Model_train_test/condition_monitoring/self_try/Joint_Monitoring_banda.py Normal file
View File

@ -0,0 +1,579 @@
# -*- coding: utf-8 -*-
# coding: utf-8
import tensorflow as tf
import tensorflow.keras
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from model.DepthwiseCon1D.DepthwiseConv1D import DepthwiseConv1D
from model.Dynamic_channelAttention.Dynamic_channelAttention import DynamicChannelAttention
from condition_monitoring.data_deal import loadData_daban as loadData
from model.Joint_Monitoring.Joint_Monitoring_banda import Joint_Monitoring
from model.CommonFunction.CommonFunction import *
from sklearn.model_selection import train_test_split
from tensorflow.keras.models import load_model, save_model
'''
@Author : dingjiawen
@Date : 2022/7/8 10:29
@Usage : 尝试将预测和分类两种方式相结合,联合监测
@Desc :REPVGG+unsampling+GRU进行重构,后面接GDP=全局动态池化+分类器
随epoch衰减的MSELoss+随epoch增强的crossEntropy
'''
'''超参数设置'''
time_stamp = 120
feature_num = 10
batch_size = 32
learning_rate = 0.001
EPOCH = 101
model_name = "banda_joint"
'''EWMA超参数'''
K = 18
namuda = 0.01
'''保存名称'''
save_name = "../hard_model/weight/{0}_epoch16_0.0009_0.0014/weight".format(model_name,
time_stamp,
feature_num,
batch_size,
EPOCH)
save_step_two_name = "../hard_model/two_weight/{0}/weight".format(model_name,
time_stamp,
feature_num,
batch_size,
EPOCH)
# save_name = "../model/joint/{0}_timestamp{1}_feature{2}.h5".format(model_name,
# time_stamp,
# feature_num,
# batch_size,
# EPOCH)
# save_step_two_name = "../model/joint_two/{0}_timestamp{1}_feature{2}.h5".format(model_name,
# time_stamp,
# feature_num,
# batch_size,
# EPOCH)
'''文件名'''
file_name = "G:\data\SCADA数据\SCADA_已处理_粤水电达坂城2020.1月-5月\风机15.csv"
'''
文件说明：jb4q_8_delete_total_zero.csv是删除了只删除了全是0的列的文件
文件从0:96748行均是正常值(2019/12.30 00:00:00 - 2020/3/11 05:58:00)
从96748:107116行均是异常值(2020/3/11 05:58:01 - 2021/3/18 11:04:00)
'''
'''文件参数'''
# 最后正常的时间点
healthy_date = 96748
# 最后异常的时间点
unhealthy_date = 107116
# 异常容忍程度
unhealthy_patience = 5
def remove(data, time_stamp=time_stamp):
rows, cols = data.shape
print("remove_data.shape:", data.shape)
num = int(rows / time_stamp)
return data[:num * time_stamp, :]
pass
# 不重叠采样
def get_training_data(data, time_stamp: int = time_stamp):
removed_data = remove(data=data)
rows, cols = removed_data.shape
print("removed_data.shape:", data.shape)
print("removed_data:", removed_data)
train_data = np.reshape(removed_data, [-1, time_stamp, cols])
print("train_data:", train_data)
batchs, time_stamp, cols = train_data.shape
for i in range(1, batchs):
each_label = np.expand_dims(train_data[i, 0, :], axis=0)
if i == 1:
train_label = each_label
else:
train_label = np.concatenate([train_label, each_label], axis=0)
print("train_data.shape:", train_data.shape)
print("train_label.shape", train_label.shape)
return train_data[:-1, :], train_label
# 重叠采样
def get_training_data_overlapping(data, time_stamp: int = time_stamp, is_Healthy: bool = True):
rows, cols = data.shape
train_data = np.empty(shape=[rows - time_stamp - 1, time_stamp, cols])
train_label = np.empty(shape=[rows - time_stamp - 1, cols])
for i in range(rows):
if i + time_stamp >= rows:
break
if i + time_stamp < rows - 1:
train_data[i] = data[i:i + time_stamp]
train_label[i] = data[i + time_stamp]
print("重叠采样以后：")
print("data:", train_data) # (300334,120,10)
print("label:", train_label) # (300334,10)
if is_Healthy:
train_label2 = np.ones(shape=[train_label.shape[0]])
else:
train_label2 = np.zeros(shape=[train_label.shape[0]])
print("label2:", train_label2)
return train_data, train_label, train_label2
# RepConv重参数化卷积
def RepConv(input_tensor, k=3):
_, _, output_dim = input_tensor.shape
conv1 = tf.keras.layers.Conv1D(filters=output_dim, kernel_size=k, strides=1, padding='SAME')(input_tensor)
b1 = tf.keras.layers.BatchNormalization()(conv1)
conv2 = tf.keras.layers.Conv1D(filters=output_dim, kernel_size=1, strides=1, padding='SAME')(input_tensor)
b2 = tf.keras.layers.BatchNormalization()(conv2)
b3 = tf.keras.layers.BatchNormalization()(input_tensor)
out = tf.keras.layers.Add()([b1, b2, b3])
out = tf.nn.relu(out)
return out
# RepBlock模块
def RepBlock(input_tensor, num: int = 3):
for i in range(num):
input_tensor = RepConv(input_tensor)
return input_tensor
# GAP 全局平均池化
def Global_avg_channelAttention(input_tensor):
_, length, channel = input_tensor.shape
DWC1 = DepthwiseConv1D(kernel_size=1, padding='SAME')(input_tensor)
GAP = tf.keras.layers.GlobalAvgPool1D()(DWC1)
c1 = tf.keras.layers.Conv1D(filters=channel, kernel_size=1, padding='SAME')(GAP)
s1 = tf.nn.sigmoid(c1)
output = tf.multiply(input_tensor, s1)
return output
# GDP 全局动态池化
def Global_Dynamic_channelAttention(input_tensor):
_, length, channel = input_tensor.shape
DWC1 = DepthwiseConv1D(kernel_size=1, padding='SAME')(input_tensor)
# GAP
GAP = tf.keras.layers.GlobalAvgPool1D()(DWC1)
c1 = tf.keras.layers.Conv1D(filters=channel, kernel_size=1, padding='SAME')(GAP)
s1 = tf.nn.sigmoid(c1)
# GMP
GMP = tf.keras.layers.GlobalMaxPool1D()(DWC1)
c2 = tf.keras.layers.Conv1D(filters=channel, kernel_size=1, padding='SAME')(GMP)
s3 = tf.nn.sigmoid(c2)
output = tf.multiply(input_tensor, s1)
return output
# 归一化
def normalization(data):
rows, cols = data.shape
print("归一化之前:", data)
print(data.shape)
print("======================")
# 归一化
max = np.max(data, axis=0)
max = np.broadcast_to(max, [rows, cols])
min = np.min(data, axis=0)
min = np.broadcast_to(min, [rows, cols])
data = (data - min) / (max - min)
print("归一化之后:", data)
print(data.shape)
return data
# 正则化
def Regularization(data):
rows, cols = data.shape
print("正则化之前:", data)
print(data.shape)
print("======================")
# 正则化
mean = np.mean(data, axis=0)
mean = np.broadcast_to(mean, shape=[rows, cols])
dst = np.sqrt(np.var(data, axis=0))
dst = np.broadcast_to(dst, shape=[rows, cols])
data = (data - mean) / dst
print("正则化之后:", data)
print(data.shape)
return data
pass
def EWMA(data, K=K, namuda=namuda):
# t是啥暂时未知
t = 0
mid = np.mean(data, axis=0)
standard = np.sqrt(np.var(data, axis=0))
UCL = mid + K * standard * np.sqrt(namuda / (2 - namuda) * (1 - (1 - namuda) ** 2 * t))
LCL = mid - K * standard * np.sqrt(namuda / (2 - namuda) * (1 - (1 - namuda) ** 2 * t))
return mid, UCL, LCL
pass
def get_MSE(data, label, new_model):
predicted_data = new_model.predict(data)
temp = np.abs(predicted_data - label)
temp1 = (temp - np.broadcast_to(np.mean(temp, axis=0), shape=predicted_data.shape))
temp2 = np.broadcast_to(np.sqrt(np.var(temp, axis=0)), shape=predicted_data.shape)
temp3 = temp1 / temp2
mse = np.sum((temp1 / temp2) ** 2, axis=1)
print("z:", mse)
print(mse.shape)
# mse=np.mean((predicted_data-label)**2,axis=1)
print("mse", mse)
dims, = mse.shape
mean = np.mean(mse)
std = np.sqrt(np.var(mse))
max = mean + 3 * std
# min = mean-3*std
max = np.broadcast_to(max, shape=[dims, ])
# min = np.broadcast_to(min,shape=[dims,])
mean = np.broadcast_to(mean, shape=[dims, ])
# plt.plot(max)
# plt.plot(mse)
# plt.plot(mean)
# # plt.plot(min)
# plt.show()
#
#
return mse, mean, max
# pass
def condition_monitoring_model():
input = tf.keras.Input(shape=[time_stamp, feature_num])
conv1 = tf.keras.layers.Conv1D(filters=256, kernel_size=1)(input)
GRU1 = tf.keras.layers.GRU(128, return_sequences=False)(conv1)
d1 = tf.keras.layers.Dense(300)(GRU1)
output = tf.keras.layers.Dense(10)(d1)
model = tf.keras.Model(inputs=input, outputs=output)
return model
# trian_data:(300455,120,10)
# trian_label1:(300455,10)
# trian_label2:(300455,)
def shuffle(train_data, train_label1, train_label2, is_split: bool = False, split_size: float = 0.2):
(train_data, test_data, train_label1, test_label1, train_label2, test_label2) = train_test_split(train_data,
train_label1,
train_label2,
test_size=split_size,
shuffle=True,
random_state=100)
if is_split:
return train_data, train_label1, train_label2, test_data, test_label1, test_label2
train_data = np.concatenate([train_data, test_data], axis=0)
train_label1 = np.concatenate([train_label1, test_label1], axis=0)
train_label2 = np.concatenate([train_label2, test_label2], axis=0)
# print(train_data.shape)
# print(train_label1.shape)
# print(train_label2.shape)
# print(train_data.shape)
return train_data, train_label1, train_label2
pass
def split_test_data(healthy_data, healthy_label1, healthy_label2, unhealthy_data, unhealthy_label1, unhealthy_label2,
split_size: float = 0.2, shuffle: bool = True):
data = np.concatenate([healthy_data, unhealthy_data], axis=0)
label1 = np.concatenate([healthy_label1, unhealthy_label1], axis=0)
label2 = np.concatenate([healthy_label2, unhealthy_label2], axis=0)
(train_data, test_data, train_label1, test_label1, train_label2, test_label2) = train_test_split(data,
label1,
label2,
test_size=split_size,
shuffle=shuffle,
random_state=100)
# print(train_data.shape)
# print(train_label1.shape)
# print(train_label2.shape)
# print(train_data.shape)
return train_data, train_label1, train_label2, test_data, test_label1, test_label2
pass
# trian_data:(300455,120,10)
# trian_label1:(300455,10)
# trian_label2:(300455,)
def train_step_one(train_data, train_label1, train_label2):
model = Joint_Monitoring()
# # # # TODO 需要运行编译一次,才能打印model.summary()
# model.build(input_shape=(batch_size, filter_num, dims))
# model.summary()
history_loss = []
history_val_loss = []
learning_rate = 1e-3
for epoch in range(EPOCH):
print()
print("EPOCH:", epoch, "/", EPOCH, ":")
train_data, train_label1, train_label2 = shuffle(train_data, train_label1, train_label2)
if epoch == 0:
train_data, train_label1, train_label2, val_data, val_label1, val_label2 = shuffle(train_data, train_label1,
train_label2,
is_split=True)
# print()
# print("EPOCH:", epoch, "/", EPOCH, ":")
# 用于让train知道，这是这个epoch中的第几次训练
z = 0
# 用于batch_size次再训练
k = 1
for data_1, label_1, label_2 in zip(train_data, train_label1, train_label2):
size, _, _ = train_data.shape
data_1 = tf.expand_dims(data_1, axis=0)
label_1 = tf.expand_dims(label_1, axis=0)
label_2 = tf.expand_dims(label_2, axis=0)
if batch_size != 1:
if k % batch_size == 1:
data = data_1
label1 = label_1
label2 = label_2
else:
data = tf.concat([data, data_1], axis=0)
label1 = tf.concat([label1, label_1], axis=0)
label2 = tf.concat([label2, label_2], axis=0)
else:
data = data_1
label1 = label_1
label2 = label_2
if k % batch_size == 0:
# label = tf.expand_dims(label, axis=-1)
loss_value, accuracy_value = model.train(input_tensor=data, label1=label1, label2=label2,
learning_rate=learning_rate,
is_first_time=True)
print(z * batch_size, "/", size, ":===============>", "loss:", loss_value.numpy())
k = 0
z = z + 1
k = k + 1
val_loss, val_accuracy = model.get_val_loss(val_data=val_data, val_label1=val_label1, val_label2=val_label2,
is_first_time=True)
SaveBestModel(model=model, save_name=save_name, history_loss=history_val_loss, loss_value=val_loss.numpy())
# SaveBestH5Model(model=model, save_name=save_name, history_loss=history_val_loss, loss_value=val_loss.numpy())
history_val_loss.append(val_loss)
history_loss.append(loss_value.numpy())
print('Training loss is :', loss_value.numpy())
print('Validating loss is :', val_loss.numpy())
if IsStopTraining(history_loss=history_val_loss, patience=7):
break
if Is_Reduce_learning_rate(history_loss=history_val_loss, patience=3):
if learning_rate >= 1e-4:
learning_rate = learning_rate * 0.1
pass
def train_step_two(step_one_model, step_two_model, train_data, train_label1, train_label2):
# step_two_model = Joint_Monitoring()
# step_two_model.build(input_shape=(batch_size, time_stamp, feature_num))
# step_two_model.summary()
history_loss = []
history_val_loss = []
history_accuracy = []
learning_rate = 1e-3
for epoch in range(EPOCH):
print()
print("EPOCH:", epoch, "/", EPOCH, ":")
train_data, train_label1, train_label2 = shuffle(train_data, train_label1, train_label2)
if epoch == 0:
train_data, train_label1, train_label2, val_data, val_label1, val_label2 = shuffle(train_data, train_label1,
train_label2,
is_split=True)
# print()
# print("EPOCH:", epoch, "/", EPOCH, ":")
# 用于让train知道，这是这个epoch中的第几次训练
z = 0
# 用于batch_size次再训练
k = 1
accuracy_num = 0
for data_1, label_1, label_2 in zip(train_data, train_label1, train_label2):
size, _, _ = train_data.shape
data_1 = tf.expand_dims(data_1, axis=0)
label_1 = tf.expand_dims(label_1, axis=0)
label_2 = tf.expand_dims(label_2, axis=0)
if batch_size != 1:
if k % batch_size == 1:
data = data_1
label1 = label_1
label2 = label_2
else:
data = tf.concat([data, data_1], axis=0)
label1 = tf.concat([label1, label_1], axis=0)
label2 = tf.concat([label2, label_2], axis=0)
else:
data = data_1
label1 = label_1
label2 = label_2
if k % batch_size == 0:
# label = tf.expand_dims(label, axis=-1)
output1, output2, output3, _ = step_one_model.call(inputs=data, is_first_time=True)
loss_value, accuracy_value = step_two_model.train(input_tensor=data, label1=label1, label2=label2,
learning_rate=learning_rate,
is_first_time=False, pred_3=output1, pred_4=output2,
pred_5=output3,epoch=epoch)
accuracy_num += accuracy_value
print(z * batch_size, "/", size, ":===============>", "loss:", loss_value.numpy(), "| accuracy:",
accuracy_num / ((z + 1) * batch_size))
k = 0
z = z + 1
k = k + 1
val_loss, val_accuracy = step_two_model.get_val_loss(val_data=val_data, val_label1=val_label1,
val_label2=val_label2,
is_first_time=False, step_one_model=step_one_model,epoch=epoch)
SaveBestModelByAccuracy(model=step_two_model, save_name=save_step_two_name, history_accuracy=history_accuracy,
accuracy_value=val_accuracy)
history_val_loss.append(val_loss)
history_loss.append(loss_value.numpy())
history_accuracy.append(val_accuracy)
print('Training loss is : {0} | Training accuracy is : {1}'.format(loss_value.numpy(),
accuracy_num / ((z + 1) * batch_size)))
print('Validating loss is : {0} | Validating accuracy is : {1}'.format(val_loss.numpy(), val_accuracy))
if IsStopTraining(history_loss=history_val_loss, patience=7):
break
if Is_Reduce_learning_rate(history_loss=history_val_loss, patience=3):
if learning_rate >= 1e-4:
learning_rate = learning_rate * 0.1
else:
print("学习率不再下降")
pass
def test(step_one_model, step_two_model, test_data, test_label1, test_label2):
history_loss = []
history_val_loss = []
val_loss, val_accuracy = step_two_model.get_val_loss(val_data=test_data, val_label1=test_label1,
val_label2=test_label2,
is_first_time=False, step_one_model=step_one_model)
history_val_loss.append(val_loss)
print("val_accuracy:", val_accuracy)
print("val_loss:", val_loss)
def showResult(step_two_model: Joint_Monitoring, test_data, isPlot: bool = False):
# 获取模型的所有参数的个数
# step_two_model.count_params()
total_result = []
size, length, dims = test_data.shape
for epoch in range(0, size - batch_size + 1, batch_size):
each_test_data = test_data[epoch:epoch + batch_size, :, :]
_, _, _, output4 = step_two_model.call(each_test_data, is_first_time=False)
total_result.append(output4)
total_result = np.reshape(total_result, [total_result.__len__(), -1])
total_result = np.reshape(total_result, [-1, ])
if isPlot:
plt.scatter(list(range(total_result.shape[0])), total_result, c='black', s=10)
# 画出 y=1 这条水平线
plt.axhline(0.5, c='red', label='Failure threshold')
# 箭头指向上面的水平线
# plt.arrow(35000, 0.9, 33000, 0.75, head_width=0.02, head_length=0.1, shape="full", fc='red', ec='red',
# alpha=0.9, overhang=0.5)
# plt.text(35000, 0.9, "Truth Fault", fontsize=10, color='black', verticalalignment='top')
plt.axvline(test_data.shape[0] * 2 / 3, c='blue', ls='-.')
plt.xlabel("time")
plt.ylabel("confience")
plt.text(total_result.shape[0] * 4 / 5, 0.6, "Fault", fontsize=10, color='black', verticalalignment='top',
horizontalalignment='center',
bbox={'facecolor': 'grey',
'pad': 10})
plt.text(total_result.shape[0] * 1 / 3, 0.4, "Norm", fontsize=10, color='black', verticalalignment='top',
horizontalalignment='center',
bbox={'facecolor': 'grey',
'pad': 10})
plt.grid()
# plt.ylim(0, 1)
# plt.xlim(-50, 1300)
# plt.legend("", loc='upper left')
plt.show()
return total_result
if __name__ == '__main__':
total_data = loadData.execute(N=feature_num, file_name=file_name)
total_data = normalization(data=total_data)
train_data_healthy, train_label1_healthy, train_label2_healthy = get_training_data_overlapping(
total_data[:healthy_date, :], is_Healthy=True)
train_data_unhealthy, train_label1_unhealthy, train_label2_unhealthy = get_training_data_overlapping(
total_data[healthy_date - time_stamp + unhealthy_patience:unhealthy_date, :],
is_Healthy=False)
#### TODO 第一步训练
# 单次测试
# train_step_one(train_data=train_data_healthy[:128, :, :], train_label1=train_label1_healthy[:128, :],train_label2=train_label2_healthy[:128, ])
# train_step_one(train_data=train_data_healthy, train_label1=train_label1_healthy, train_label2=train_label2_healthy)
# 导入第一步已经训练好的模型,一个继续训练，一个只输出结果
step_one_model = Joint_Monitoring()
step_one_model.load_weights(save_name)
step_two_model = Joint_Monitoring()
step_two_model.load_weights(save_name)
#### TODO 第二步训练
### healthy_data.shape: (300333,120,10)
### unhealthy_data.shape: (16594,10)
healthy_size, _, _ = train_data_healthy.shape
unhealthy_size, _, _ = train_data_unhealthy.shape
train_data, train_label1, train_label2, test_data, test_label1, test_label2 = split_test_data(
healthy_data=train_data_healthy[healthy_size - 2 * unhealthy_size:, :, :],
healthy_label1=train_label1_healthy[healthy_size - 2 * unhealthy_size:, :],
healthy_label2=train_label2_healthy[healthy_size - 2 * unhealthy_size:, ], unhealthy_data=train_data_unhealthy,
unhealthy_label1=train_label1_unhealthy, unhealthy_label2=train_label2_unhealthy)
train_step_two(step_one_model=step_one_model, step_two_model=step_two_model,
train_data=train_data,
train_label1=train_label1, train_label2=np.expand_dims(train_label2, axis=-1))
### TODO 测试测试集
step_one_model = Joint_Monitoring()
step_one_model.load_weights(save_name)
step_two_model = Joint_Monitoring()
step_two_model.load_weights(save_step_two_name)
# test(step_one_model=step_one_model, step_two_model=step_two_model, test_data=test_data, test_label1=test_label1,
# test_label2=np.expand_dims(test_label2, axis=-1))
###TODO 展示全部的结果
all_data, _, _ = get_training_data_overlapping(
total_data[healthy_size - 2 * unhealthy_size:unhealthy_date, :], is_Healthy=True)
# all_data = np.concatenate([])
# 单次测试
# showResult(step_two_model, test_data=all_data[:32], isPlot=True)
showResult(step_two_model, test_data=all_data, isPlot=True)
pass

25
TensorFlow_eaxmple/Model_train_test/condition_monitoring/self_try/compare/RNet-L.py
View File

@ -17,7 +17,8 @@ import pandas as pd
import matplotlib.pyplot as plt
from model.DepthwiseCon1D.DepthwiseConv1D import DepthwiseConv1D
from model.Dynamic_channelAttention.Dynamic_channelAttention import DynamicChannelAttention
from condition_monitoring.data_deal import loadData
from condition_monitoring.data_deal import loadData_daban as loadData
from model.Joint_Monitoring.compare.RNet_L import Joint_Monitoring
from model.CommonFunction.CommonFunction import *
@ -28,16 +29,16 @@ import random
'''超参数设置'''
time_stamp = 120
feature_num = 10
batch_size = 16
batch_size = 32
learning_rate = 0.001
EPOCH = 101
model_name = "RNet_L"
model_name = "RNet_L_banda"
'''EWMA超参数'''
K = 18
namuda = 0.01
'''保存名称'''
save_name = "./model/weight/{0}_timestamp{1}_feature{2}_weight/weight".format(model_name,
save_name = "./model/weight/{0}/weight".format(model_name,
time_stamp,
feature_num,
batch_size,
@ -70,7 +71,7 @@ save_max_name = "./mse/{0}/{0}_timestamp{1}_feature{2}_max.csv".format(model_nam
# batch_size,
# EPOCH)
'''文件名'''
file_name = "G:\data\SCADA数据\jb4q_8_delete_total_zero.csv"
file_name = "G:\data\SCADA数据\SCADA_已处理_粤水电达坂城2020.1月-5月\风机15.csv"
'''
文件说明：jb4q_8_delete_total_zero.csv是删除了只删除了全是0的列的文件
@ -79,9 +80,9 @@ file_name = "G:\data\SCADA数据\jb4q_8_delete_total_zero.csv"
'''
'''文件参数'''
# 最后正常的时间点
healthy_date = 415548
healthy_date = 96748
# 最后异常的时间点
unhealthy_date = 432153
unhealthy_date = 107116
# 异常容忍程度
unhealthy_patience = 5
@ -360,6 +361,8 @@ def train_step_one(train_data, train_label1, train_label2):
if Is_Reduce_learning_rate(history_loss=history_val_loss, patience=3):
if learning_rate >= 1e-4:
learning_rate = learning_rate * 0.1
else:
print("学习率不再下降")
pass
@ -435,6 +438,8 @@ def train_step_two(step_one_model, step_two_model, train_data, train_label1, tra
if Is_Reduce_learning_rate(history_loss=history_val_loss, patience=3):
if learning_rate >= 1e-4:
learning_rate = learning_rate * 0.1
else:
print("学习率不再下降")
pass
@ -624,8 +629,8 @@ def getResult(model: tf.keras.Model, healthy_data, healthy_label, unhealthy_data
if __name__ == '__main__':
# total_data = loadData.execute(N=feature_num, file_name=file_name)
total_data=np.load("G:\data\SCADA数据\靖边8号处理后的数据\原始10SCADA数据/total_data.npy")
total_data = loadData.execute(N=feature_num, file_name=file_name)
# total_data=np.load("G:\data\SCADA数据\靖边8号处理后的数据\原始10SCADA数据/total_data.npy")
total_data = normalization(data=total_data)
train_data_healthy, train_label1_healthy, train_label2_healthy = get_training_data_overlapping(
total_data[:healthy_date, :], is_Healthy=True)
@ -637,7 +642,7 @@ if __name__ == '__main__':
# train_step_one(train_data=train_data_healthy[:256, :, :], train_label1=train_label1_healthy[:256, :],
# train_label2=train_label2_healthy[:256, ])
#### 模型训练
# train_step_one(train_data=train_data_healthy, train_label1=train_label1_healthy, train_label2=train_label2_healthy)
train_step_one(train_data=train_data_healthy, train_label1=train_label1_healthy, train_label2=train_label2_healthy)
# 导入第一步已经训练好的模型,一个继续训练，一个只输出结果
step_one_model = Joint_Monitoring()

13
TensorFlow_eaxmple/Model_train_test/condition_monitoring/self_try/compare/RNet.py
View File

@ -27,16 +27,16 @@ import random
'''超参数设置'''
time_stamp = 120
feature_num = 10
batch_size = 16
batch_size = 32
learning_rate = 0.001
EPOCH = 101
model_name = "RNet"
model_name = "RNet_banda"
'''EWMA超参数'''
K = 18
namuda = 0.01
'''保存名称'''
save_name = "./model/weight/{0}_timestamp{1}_feature{2}_weight_epoch2_loss0.007/weight".format(model_name,
save_name = "./model/weight/{0}/weight".format(model_name,
time_stamp,
feature_num,
batch_size,
@ -58,7 +58,8 @@ save_step_two_name = "./model/two_weight/{0}_timestamp{1}_feature{2}_weight/weig
# batch_size,
# EPOCH)
'''文件名'''
file_name = "G:\data\SCADA数据\jb4q_8_delete_total_zero.csv"
'''文件名'''
file_name = "G:\data\SCADA数据\SCADA_已处理_粤水电达坂城2020.1月-5月\风机15.csv"
'''
文件说明：jb4q_8_delete_total_zero.csv是删除了只删除了全是0的列的文件
@ -67,9 +68,9 @@ file_name = "G:\data\SCADA数据\jb4q_8_delete_total_zero.csv"
'''
'''文件参数'''
# 最后正常的时间点
healthy_date = 415548
healthy_date = 96748
# 最后异常的时间点
unhealthy_date = 432153
unhealthy_date = 107116
# 异常容忍程度
unhealthy_patience = 5

460
TensorFlow_eaxmple/Model_train_test/model/Joint_Monitoring/Joint_Monitoring_banda.py Normal file
View File

@ -0,0 +1,460 @@
# _*_ coding: UTF-8 _*_
'''
@Author : dingjiawen
@Date : 2022/7/14 9:40
@Usage : 联合监测模型
@Desc : 将预测值放入分类器,分类器放两层逐渐递减的dense层
'''
import tensorflow as tf
import tensorflow.keras as keras
from tensorflow.keras import *
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from model.DepthwiseCon1D.DepthwiseConv1D import DepthwiseConv1D
from model.Dynamic_channelAttention.Dynamic_channelAttention import DynamicChannelAttention, DynamicPooling
from condition_monitoring.data_deal import loadData
from model.LossFunction.smooth_L1_Loss import SmoothL1Loss
import math
class Joint_Monitoring(keras.Model):
def __init__(self, conv_filter=20,epochs=50):
# 调用父类__init__()方法
super(Joint_Monitoring, self).__init__()
# step one
self.RepDCBlock1 = RevConvBlock(num=3, kernel_size=5)
self.conv1 = tf.keras.layers.Conv1D(filters=conv_filter, kernel_size=1, strides=2, padding='SAME')
self.upsample1 = tf.keras.layers.UpSampling1D(size=2)
self.DACU2 = DynamicChannelAttention()
self.RepDCBlock2 = RevConvBlock(num=3, kernel_size=3)
self.conv2 = tf.keras.layers.Conv1D(filters=2 * conv_filter, kernel_size=1, strides=2, padding='SAME')
self.upsample2 = tf.keras.layers.UpSampling1D(size=2)
self.DACU3 = DynamicChannelAttention()
self.RepDCBlock3 = RevConvBlock(num=3, kernel_size=3)
self.p1 = DynamicPooling(pool_size=2)
self.conv3 = tf.keras.layers.Conv1D(filters=2 * conv_filter, kernel_size=3, strides=2, padding='SAME')
self.DACU4 = DynamicChannelAttention()
self.RepDCBlock4 = RevConvBlock(num=3, kernel_size=3)
self.p2 = DynamicPooling(pool_size=4)
self.conv4 = tf.keras.layers.Conv1D(filters=conv_filter, kernel_size=3, strides=2, padding='SAME')
self.RepDCBlock5 = RevConvBlock(num=3, kernel_size=3)
self.p3 = DynamicPooling(pool_size=2)
# step two
# 重现原数据
self.GRU1 = tf.keras.layers.GRU(128, return_sequences=False)
self.d1 = tf.keras.layers.Dense(300, activation=tf.nn.leaky_relu)
self.output1 = tf.keras.layers.Dense(10, activation=tf.nn.leaky_relu)
self.GRU2 = tf.keras.layers.GRU(128, return_sequences=False)
self.d2 = tf.keras.layers.Dense(300, activation=tf.nn.leaky_relu)
self.output2 = tf.keras.layers.Dense(10, activation=tf.nn.leaky_relu)
self.GRU3 = tf.keras.layers.GRU(128, return_sequences=False)
self.d3 = tf.keras.layers.Dense(300, activation=tf.nn.leaky_relu)
self.output3 = tf.keras.layers.Dense(10, activation=tf.nn.leaky_relu)
# step three
# 分类器
self.d4 = tf.keras.layers.Dense(300, activation=tf.nn.leaky_relu)
self.d5 = tf.keras.layers.Dense(10, activation=tf.nn.leaky_relu)
# tf.nn.softmax
self.output4 = tf.keras.layers.Dense(1, activation=tf.nn.sigmoid)
# loss
self.train_loss = []
self.epoch=0
self.epochs=epochs
def call(self, inputs, training=None, mask=None, is_first_time: bool = True):
# step one
RepDCBlock1 = self.RepDCBlock1(inputs)
RepDCBlock1 = tf.keras.layers.BatchNormalization()(RepDCBlock1)
conv1 = self.conv1(RepDCBlock1)
conv1 = tf.nn.leaky_relu(conv1)
conv1 = tf.keras.layers.BatchNormalization()(conv1)
upsample1 = self.upsample1(conv1)
DACU2 = self.DACU2(upsample1)
DACU2 = tf.keras.layers.BatchNormalization()(DACU2)
RepDCBlock2 = self.RepDCBlock2(DACU2)
RepDCBlock2 = tf.keras.layers.BatchNormalization()(RepDCBlock2)
conv2 = self.conv2(RepDCBlock2)
conv2 = tf.nn.leaky_relu(conv2)
conv2 = tf.keras.layers.BatchNormalization()(conv2)
upsample2 = self.upsample2(conv2)
DACU3 = self.DACU3(upsample2)
DACU3 = tf.keras.layers.BatchNormalization()(DACU3)
RepDCBlock3 = self.RepDCBlock3(DACU3)
RepDCBlock3 = tf.keras.layers.BatchNormalization()(RepDCBlock3)
conv3 = self.conv3(RepDCBlock3)
conv3 = tf.nn.leaky_relu(conv3)
conv3 = tf.keras.layers.BatchNormalization()(conv3)
concat1 = tf.concat([conv2, conv3], axis=1)
DACU4 = self.DACU4(concat1)
DACU4 = tf.keras.layers.BatchNormalization()(DACU4)
RepDCBlock4 = self.RepDCBlock4(DACU4)
RepDCBlock4 = tf.keras.layers.BatchNormalization()(RepDCBlock4)
conv4 = self.conv4(RepDCBlock4)
conv4 = tf.nn.leaky_relu(conv4)
conv4 = tf.keras.layers.BatchNormalization()(conv4)
concat2 = tf.concat([conv1, conv4], axis=1)
RepDCBlock5 = self.RepDCBlock5(concat2)
RepDCBlock5 = tf.keras.layers.BatchNormalization()(RepDCBlock5)
output1 = []
output2 = []
output3 = []
output4 = []
if is_first_time:
# step two
# 重现原数据
# 接block3
GRU1 = self.GRU1(RepDCBlock3)
GRU1 = tf.keras.layers.BatchNormalization()(GRU1)
d1 = self.d1(GRU1)
# tf.nn.softmax
output1 = self.output1(d1)
# 接block4
GRU2 = self.GRU2(RepDCBlock4)
GRU2 = tf.keras.layers.BatchNormalization()(GRU2)
d2 = self.d2(GRU2)
# tf.nn.softmax
output2 = self.output2(d2)
# 接block5
GRU3 = self.GRU3(RepDCBlock5)
GRU3 = tf.keras.layers.BatchNormalization()(GRU3)
d3 = self.d3(GRU3)
# tf.nn.softmax
output3 = self.output3(d3)
else:
GRU1 = self.GRU1(RepDCBlock3)
GRU1 = tf.keras.layers.BatchNormalization()(GRU1)
d1 = self.d1(GRU1)
# tf.nn.softmax
output1 = self.output1(d1)
# 接block4
GRU2 = self.GRU2(RepDCBlock4)
GRU2 = tf.keras.layers.BatchNormalization()(GRU2)
d2 = self.d2(GRU2)
# tf.nn.softmax
output2 = self.output2(d2)
# 接block5
GRU3 = self.GRU3(RepDCBlock5)
GRU3 = tf.keras.layers.BatchNormalization()(GRU3)
d3 = self.d3(GRU3)
# tf.nn.softmax
output3 = self.output3(d3)
# 多尺度动态池化
# p1 = self.p1(output1)
# B, _, _ = p1.shape
# f1 = tf.reshape(p1, shape=[B, -1])
# p2 = self.p2(output2)
# f2 = tf.reshape(p2, shape=[B, -1])
# p3 = self.p3(output3)
# f3 = tf.reshape(p3, shape=[B, -1])
# step three
# 分类器
concat3 = tf.concat([output1, output2, output3], axis=1)
# dropout = tf.keras.layers.Dropout(0.25)(concat3)
d4 = self.d4(concat3)
d5 = self.d5(d4)
# d4 = tf.keras.layers.BatchNormalization()(d4)
output4 = self.output4(d5)
return output1, output2, output3, output4
def get_loss(self, inputs_tensor, label1=None, label2=None, is_first_time: bool = True, pred_3=None, pred_4=None,
pred_5=None):
# step one
RepDCBlock1 = self.RepDCBlock1(inputs_tensor)
RepDCBlock1 = tf.keras.layers.BatchNormalization()(RepDCBlock1)
conv1 = self.conv1(RepDCBlock1)
conv1 = tf.nn.leaky_relu(conv1)
conv1 = tf.keras.layers.BatchNormalization()(conv1)
upsample1 = self.upsample1(conv1)
DACU2 = self.DACU2(upsample1)
DACU2 = tf.keras.layers.BatchNormalization()(DACU2)
RepDCBlock2 = self.RepDCBlock2(DACU2)
RepDCBlock2 = tf.keras.layers.BatchNormalization()(RepDCBlock2)
conv2 = self.conv2(RepDCBlock2)
conv2 = tf.nn.leaky_relu(conv2)
conv2 = tf.keras.layers.BatchNormalization()(conv2)
upsample2 = self.upsample2(conv2)
DACU3 = self.DACU3(upsample2)
DACU3 = tf.keras.layers.BatchNormalization()(DACU3)
RepDCBlock3 = self.RepDCBlock3(DACU3)
RepDCBlock3 = tf.keras.layers.BatchNormalization()(RepDCBlock3)
conv3 = self.conv3(RepDCBlock3)
conv3 = tf.nn.leaky_relu(conv3)
conv3 = tf.keras.layers.BatchNormalization()(conv3)
concat1 = tf.concat([conv2, conv3], axis=1)
DACU4 = self.DACU4(concat1)
DACU4 = tf.keras.layers.BatchNormalization()(DACU4)
RepDCBlock4 = self.RepDCBlock4(DACU4)
RepDCBlock4 = tf.keras.layers.BatchNormalization()(RepDCBlock4)
conv4 = self.conv4(RepDCBlock4)
conv4 = tf.nn.leaky_relu(conv4)
conv4 = tf.keras.layers.BatchNormalization()(conv4)
concat2 = tf.concat([conv1, conv4], axis=1)
RepDCBlock5 = self.RepDCBlock5(concat2)
RepDCBlock5 = tf.keras.layers.BatchNormalization()(RepDCBlock5)
if is_first_time:
# step two
# 重现原数据
# 接block3
GRU1 = self.GRU1(RepDCBlock3)
GRU1 = tf.keras.layers.BatchNormalization()(GRU1)
d1 = self.d1(GRU1)
# tf.nn.softmax
output1 = self.output1(d1)
# 接block4
GRU2 = self.GRU2(RepDCBlock4)
GRU2 = tf.keras.layers.BatchNormalization()(GRU2)
d2 = self.d2(GRU2)
# tf.nn.softmax
output2 = self.output2(d2)
# 接block5
GRU3 = self.GRU3(RepDCBlock5)
GRU3 = tf.keras.layers.BatchNormalization()(GRU3)
d3 = self.d3(GRU3)
# tf.nn.softmax
output3 = self.output3(d3)
# reduce_mean降维计算均值
MSE_loss1 = SmoothL1Loss()(y_true=label1, y_pred=output1)
MSE_loss2 = SmoothL1Loss()(y_true=label1, y_pred=output2)
MSE_loss3 = SmoothL1Loss()(y_true=label1, y_pred=output3)
# MSE_loss1 = tf.reduce_mean(tf.keras.losses.mse(y_true=label1, y_pred=output1))
# MSE_loss2 = tf.reduce_mean(tf.keras.losses.mse(y_true=label1, y_pred=output2))
# MSE_loss3 = tf.reduce_mean(tf.keras.losses.mse(y_true=label1, y_pred=output3))
print("MSE_loss1:", MSE_loss1.numpy())
print("MSE_loss2:", MSE_loss2.numpy())
print("MSE_loss3:", MSE_loss3.numpy())
loss = MSE_loss1 + MSE_loss2 + MSE_loss3
Accuracy_num = 0
else:
# step two
# 重现原数据
# 接block3
GRU1 = self.GRU1(RepDCBlock3)
GRU1 = tf.keras.layers.BatchNormalization()(GRU1)
d1 = self.d1(GRU1)
# tf.nn.softmax
output1 = self.output1(d1)
# 接block4
GRU2 = self.GRU2(RepDCBlock4)
GRU2 = tf.keras.layers.BatchNormalization()(GRU2)
d2 = self.d2(GRU2)
# tf.nn.softmax
output2 = self.output2(d2)
# 接block5
GRU3 = self.GRU3(RepDCBlock5)
GRU3 = tf.keras.layers.BatchNormalization()(GRU3)
d3 = self.d3(GRU3)
# tf.nn.softmax
output3 = self.output3(d3)
# 多尺度动态池化
# p1 = self.p1(output1)
# B, _, _ = p1.shape
# f1 = tf.reshape(p1, shape=[B, -1])
# p2 = self.p2(output2)
# f2 = tf.reshape(p2, shape=[B, -1])
# p3 = self.p3(output3)
# f3 = tf.reshape(p3, shape=[B, -1])
# step three
# 分类器
concat3 = tf.concat([output1, output2, output3], axis=1)
# dropout = tf.keras.layers.Dropout(0.25)(concat3)
d4 = self.d4(concat3)
d5 = self.d5(d4)
# d4 = tf.keras.layers.BatchNormalization()(d4)
output4 = self.output4(d5)
# reduce_mean降维计算均值
a = 0.02
beta = 0.5 * math.cos(min(self.epoch * 2 / self.epochs, 1) * math.pi) + 0.5
MSE_loss = SmoothL1Loss()(y_true=pred_3, y_pred=output1)
MSE_loss += SmoothL1Loss()(y_true=pred_4, y_pred=output2)
MSE_loss += SmoothL1Loss()(y_true=pred_5, y_pred=output3)
Cross_Entropy_loss = tf.reduce_mean(
tf.losses.binary_crossentropy(y_true=label2, y_pred=output4, from_logits=True))
print("MSE_loss:", MSE_loss.numpy())
print("Cross_Entropy_loss:", Cross_Entropy_loss.numpy())
Accuracy_num = self.get_Accuracy(label=label2, output=output4)
loss = beta * MSE_loss + a * Cross_Entropy_loss
return loss, Accuracy_num
def get_Accuracy(self, output, label):
predict_label = tf.round(output)
label = tf.cast(label, dtype=tf.float32)
t = np.array(label - predict_label)
b = t[t[:] == 0]
return b.__len__()
def get_grad(self, input_tensor, label1=None, label2=None, is_first_time: bool = True, pred_3=None, pred_4=None,
pred_5=None):
with tf.GradientTape() as tape:
# todo 原本tape只会监控由tf.Variable创建的trainable=True属性
# tape.watch(self.variables)
L, Accuracy_num = self.get_loss(input_tensor, label1=label1, label2=label2, is_first_time=is_first_time,
pred_3=pred_3,
pred_4=pred_4, pred_5=pred_5)
# 保存一下loss，用于输出
self.train_loss = L
g = tape.gradient(L, self.variables)
return g, Accuracy_num
def train(self, input_tensor, label1=None, label2=None, learning_rate=1e-3, is_first_time: bool = True, pred_3=None,
pred_4=None, pred_5=None,epoch=0):
self.epoch=epoch
g, Accuracy_num = self.get_grad(input_tensor, label1=label1, label2=label2, is_first_time=is_first_time,
pred_3=pred_3,
pred_4=pred_4, pred_5=pred_5)
optimizers.Adam(learning_rate).apply_gradients(zip(g, self.variables))
return self.train_loss, Accuracy_num
# 暂时只支持batch_size等于1,不然要传z比较麻烦
def get_val_loss(self, val_data, val_label1, val_label2, batch_size=16, is_first_time: bool = True,
step_one_model=None,epoch=0):
val_loss = []
self.epoch=epoch
accuracy_num = 0
output1 = 0
output2 = 0
output3 = 0
z = 1
size, length, dims = val_data.shape
if batch_size == None:
batch_size = self.batch_size
for epoch in range(0, size - batch_size, batch_size):
each_val_data = val_data[epoch:epoch + batch_size, :, :]
each_val_label1 = val_label1[epoch:epoch + batch_size, :]
each_val_label2 = val_label2[epoch:epoch + batch_size, ]
# each_val_data = tf.expand_dims(each_val_data, axis=0)
# each_val_query = tf.expand_dims(each_val_query, axis=0)
# each_val_label = tf.expand_dims(each_val_label, axis=0)
if not is_first_time:
output1, output2, output3, _ = step_one_model.call(inputs=each_val_data, is_first_time=True)
each_loss, each_accuracy_num = self.get_loss(each_val_data, each_val_label1, each_val_label2,
is_first_time=is_first_time,
pred_3=output1, pred_4=output2, pred_5=output3)
accuracy_num += each_accuracy_num
val_loss.append(each_loss)
z += 1
val_accuracy = accuracy_num / ((z - 1) * batch_size)
val_total_loss = tf.reduce_mean(val_loss)
return val_total_loss, val_accuracy
class RevConv(keras.layers.Layer):
def __init__(self, kernel_size=3):
# 调用父类__init__()方法
super(RevConv, self).__init__()
self.kernel_size = kernel_size
def get_config(self):
# 自定义层里面的属性
config = (
{
'kernel_size': self.kernel_size
}
)
base_config = super(RevConv, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
def build(self, input_shape):
# print(input_shape)
_, _, output_dim = input_shape[0], input_shape[1], input_shape[2]
self.conv1 = tf.keras.layers.Conv1D(filters=output_dim, kernel_size=self.kernel_size, strides=1,
padding='causal',
dilation_rate=4)
self.conv2 = tf.keras.layers.Conv1D(filters=output_dim, kernel_size=1, strides=1, padding='causal',
dilation_rate=4)
# self.b2 = tf.keras.layers.BatchNormalization()
# self.b3 = tf.keras.layers.BatchNormalization()
# out = tf.keras.layers.Add()([b1, b2, b3])
# out = tf.nn.relu(out)
def call(self, inputs, **kwargs):
conv1 = self.conv1(inputs)
b1 = tf.keras.layers.BatchNormalization()(conv1)
b1 = tf.nn.leaky_relu(b1)
# b1 = self.b1
conv2 = self.conv2(inputs)
b2 = tf.keras.layers.BatchNormalization()(conv2)
b2 = tf.nn.leaky_relu(b2)
b3 = tf.keras.layers.BatchNormalization()(inputs)
out = tf.keras.layers.Add()([b1, b2, b3])
out = tf.nn.relu(out)
return out
class RevConvBlock(keras.layers.Layer):
def __init__(self, num: int = 3, kernel_size=3):
# 调用父类__init__()方法
super(RevConvBlock, self).__init__()
self.num = num
self.kernel_size = kernel_size
self.L = []
for i in range(num):
RepVGG = RevConv(kernel_size=kernel_size)
self.L.append(RepVGG)
def get_config(self):
# 自定义层里面的属性
config = (
{
'kernel_size': self.kernel_size,
'num': self.num
}
)
base_config = super(RevConvBlock, self).get_config()
return dict(list(base_config.items()) + list(config.items()))
def call(self, inputs, **kwargs):
for i in range(self.num):
inputs = self.L[i](inputs)
return inputs