-
Notifications
You must be signed in to change notification settings - Fork 0
/
Visualization_Utils.py
200 lines (143 loc) · 5.34 KB
/
Visualization_Utils.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
# -*- coding: utf-8 -*-
"""
Created on Sun Sep 12 11:09:16 2021
@author: srpv
"""
import torch
import copy
import numpy as np
import pandas as pd
import seaborn as sns
# from pylab import rcParams
import matplotlib.pyplot as plt
from matplotlib import rc
from sklearn.model_selection import train_test_split
from torch.optim.lr_scheduler import StepLR
from torch import nn, optim
import torch.nn.functional as F
import matplotlib.patches as mpatches
from matplotlib import cm
from Utils import *
from tSNE import *
# rcParams['figure.figsize'] = 12, 8
RANDOM_SEED = 42
np.random.seed(RANDOM_SEED)
torch.manual_seed(RANDOM_SEED)
def path_leaf(path):
head, tail = ntpath.split(path)
return tail
def reconstruction_loss(model, force_df):
plt.rcParams.update(plt.rcParamsDefault)
tail=path_leaf(force_df)
A=tail.split('.')
A=A[0]
print(A)
filename='Reconstruction_loss_distribution_'+str(A)+'.png'
filename2='Reconstruction_loss_Histogram_distribution'+str(A)+'.png'
force_df = np.load(force_df)
force_df = pd.DataFrame(force_df)
print(force_df.shape)
force_dataset, seq_len, n_features = create_dataset(force_df)
_,_,force_losses = predict(model, force_dataset)
losses = np.array(force_losses)
windowsize= len(losses) #window size you wish to visualise
sample_rate= 1
dt=1/sample_rate
t0=0
time = np.arange(0, windowsize) * dt + t0
time = np.expand_dims(time, axis=1)
losses=losses.squeeze()
time=time.squeeze()
fig, ax = plt.subplots(figsize=(10,6))
plt.vlines(time,ymin=0, ymax=losses, color='blue', alpha=1,label="Reconstruction error")
plt.ylabel('Reconstruction error',fontsize=15)
plt.xlabel('cycle no',fontsize=15)
plt.ticklabel_format(axis='x', style='sci',scilimits=(0,0))
plt.ticklabel_format(axis='y', style='sci',scilimits=(0,0))
plt.savefig(filename,dpi=800)
plt.show()
plt.clf()
#%%
def reconstruction_loss_scatter(model, force_df,THRESHOLD,n1,n2):
tail=path_leaf(force_df)
A=tail.split('.')
A=A[0]
print(A)
filename1='Error_scatter_distribution_'+str(A)+'.png'
filename2='Error_scatter_bar_'+str(A)+'.png'
force_df = np.load(force_df)
force_df = pd.DataFrame(force_df)
print(force_df.shape)
force_dataset, seq_len, n_features = create_dataset(force_df)
_,_,force_losses = predict(model, force_dataset)
losses = np.array(force_losses)
anomaly_indices = []
anomaly_values = []
error_values = []
for i in range(losses.shape[0]):
error=losses[i]-THRESHOLD
error_values.append(error)
if i > n2 or i < n1 :
# print(i)
if losses[i] >= THRESHOLD:
anomaly_indices.append(i)
anomaly_values.append(losses[i])
plt.rcParams.update(plt.rcParamsDefault)
fig, ax = plt.subplots(figsize=(10,6))
ax.plot(losses, label="Reconstruction error",color='blue',linewidth=0.75)
plt.scatter(x=anomaly_indices, y=anomaly_values, color='red', label="Anomaly")
plt.axhline(y=THRESHOLD, c='r', linestyle='--',linewidth=0.5)
plt.legend(loc = "upper left")
plt.savefig(filename1,dpi=800)
plt.show()
plt.clf()
losses = np.array(force_losses)
losses = pd.DataFrame(losses)
losses=losses.set_axis(['Reconstruction_error'], axis=1)
plt.rcParams.update(plt.rcParamsDefault)
fig, ax = plt.subplots(figsize=(10,6))
ax = losses.plot(kind='bar',color='blue')
plt.scatter(x=anomaly_indices, y=anomaly_values, color='red', label="Anomaly")
plt.axhline(y=THRESHOLD, c='r', linestyle='--',linewidth=0.5)
plt.legend(loc = "upper left")
ax.locator_params(nbins=10, axis='x')
plt.ylabel('Reconstruction error',fontsize=15)
plt.xlabel('Cycle no',fontsize=15)
plt.title('Reconstruction error over cycle',fontsize=20)
plt.savefig(filename2,dpi=800)
plt.show()
plt.clf()
#%%
def plotsignal(li,color,filename,add):
plt.rcParams.update(plt.rcParamsDefault)
windowsize= len(li) #window size you wish to visualise
sample_rate= 100
dt=1/sample_rate
t0=0
time = np.arange(0, windowsize) * dt + t0
time=time+add
fig, ax = plt.subplots(figsize=(10, 3))
plt.plot(time,li,color=color, linewidth=3)
plt.ylabel('Amplitude (V)',fontsize=15)
plt.xlabel('cycle no',fontsize=15)
plotname='Force_cycle_'+str(filename)+'.png'
plt.savefig(plotname,bbox_inches='tight',pad_inches=0.1,dpi=200)
plt.show()
plt.clf()
def plotsignals(Running_in,Stable,Anomaly):
plt.rcParams.update(plt.rcParamsDefault)
windowsize= len(Running_in)
sample_rate= 100
dt=1/sample_rate
t0=0
time = np.arange(0, windowsize) * dt + t0
fig, ax = plt.subplots(figsize=(10, 3))
plt.plot(time,Running_in,color="#FF1818", linewidth=2,label='Running_in')
plt.plot(time,Stable,color="#00CB26", linewidth=2,label='Stable')
plt.plot(time,Anomaly,color="#0000EC", linewidth=2,label='Anomaly')
plt.ylabel('Amplitude (V)',fontsize=15)
plt.xlabel('cycle no',fontsize=15)
plotname='Force_cycle_comparison'+'.png'
plt.legend(loc='upper right',bbox_to_anchor=(1.20, 1))
plt.savefig(plotname,bbox_inches='tight',pad_inches=0.1,dpi=200)
plt.show()