-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathTst_CNN_predictor_accuracy.py
154 lines (141 loc) · 6.7 KB
/
Tst_CNN_predictor_accuracy.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
import numpy as np
import torch
from torch import nn
from torch.utils.data import DataLoader
from MyDataLoader import MyNoiseDataset
from Tst_CNN_predicotr_v1 import Filter_ID_predictor, Filter_ID_predictor_from_1DCNN_LMSoftmax
from Train_validate import create_data_loader
from sklearn.metrics import classification_report, confusion_matrix, ConfusionMatrixDisplay
from Bcolors import bcolors
import matplotlib.pyplot as plt
#----------------------------------------------------------------
from pytorch_metric_learning import losses
from pytorch_metric_learning.utils.inference import LogitGetter
#-------------------------------------------------------------
# Function : load_weigth_for_model()
# Loading the weights to model from pre-trained coefficients
#-------------------------------------------------------------
def load_weigth_for_model(model, pretrained_path, device):
model_dict = model.state_dict()
pretrained_dict = torch.load(pretrained_path,map_location= device)
for k, v in model_dict.items():
model_dict[k] = pretrained_dict[k]
model.load_state_dict(model_dict)
#-----------------------------------------------------------------
# Function :
# Description :
#-----------------------------------------------------------------
def tst_accuracy_of_model(tst_data_loder, model, return_index =None): #return_index = None
if return_index == None :
accuracy_vec = []
i = 0
for input, target in tst_data_loder:
i += 1
batch_acc = 0
for signal_1d, target_1d in zip(input, target):
batch_acc +=(model.predic_ID(signal_1d)==target_1d.numpy())
acc = batch_acc/len(target)
print(f"----------------------------------------")
print(f"The {i}-th iteration's accuracy is {acc}")
accuracy_vec.append(acc)
return accuracy_vec, sum(accuracy_vec)/len(accuracy_vec)
else:
i = 0
predict_index = []
target_index = []
accuracy_vec = []
for input, target in tst_data_loder:
i += 1
batch_acc = 0
for signal_1d, target_1d in zip(input, target):
pre, tar = model.predic_ID(signal_1d), target_1d.numpy()
predict_index.append(pre)
target_index.append(tar)
batch_acc += (pre == tar)
acc = batch_acc/len(target)
print(f"----------------------------------------")
print(f"The {i}-th iteration's accuracy is {acc}")
accuracy_vec.append(acc)
return accuracy_vec, sum(accuracy_vec)/len(accuracy_vec), predict_index, target_index
#----------------------------------------------------------------
# Function : Testing accuracy of the predictor (coming from main)
#----------------------------------------------------------------
def Testing_model_accuracy(MODEL_PATH, MATFILE_PATH, VALIDATTION_FILE, Report=None, Class_Num=5):
if torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
print(f"Using {device}")
#D
BATCH_SIZE = 100
fs = 16000
sheet = "Index.csv"
CNN_classfier = Filter_ID_predictor(MODEL_PATH, MATFILE_PATH, fs, device)
valid_data = MyNoiseDataset(VALIDATTION_FILE,sheet)
valid_dataloader = create_data_loader(valid_data,BATCH_SIZE)
if Report == None:
_, average_acc = tst_accuracy_of_model(valid_dataloader,CNN_classfier)
print(f"The average accuracy is {average_acc}")
else:
_, average_acc, y_pred, y_true = tst_accuracy_of_model(valid_dataloader,CNN_classfier,return_index=True)
print(bcolors.OKCYAN + f"The average accuracy is {average_acc}" + bcolors.ENDC)
target_names = []
for jj in range(Class_Num):
target_names.append(f'class {jj}')
#target_names = ['class 0', 'class 1', 'class 2','class 3', 'class 4']
print(bcolors.RED + '<<====================Classification report=========================>>' + bcolors.ENDC)
print(classification_report(y_true, y_pred, target_names=target_names))
print(bcolors.RED +'<<==========================End=====================================>>' + bcolors.ENDC)
cm = confusion_matrix(y_true, y_pred)
disp = ConfusionMatrixDisplay(confusion_matrix=cm,display_labels=target_names)
disp.plot()
plt.show()
return average_acc
#------------------------------------------------------------------------------------------
# Function : Testing accuracy of the predictor of 1D_CNN with LMSoftmax (coming from main)
#------------------------------------------------------------------------------------------
def Testing_model_with_LMSoftmax_accuracy(MODEL_PATH,MATFILE_PATH, VALIDATTION_FILE, LMSOFTMAX_WEIGHT_PTH):
if torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
print(f"Using {device}")
#
BATCH_SIZE = 100
fs = 16000
sheet = "Index.csv"
LMSoftmax_weight =losses.LargeMarginSoftmaxLoss(num_classes= 15, embedding_size= 620, margin = 2).to(device)
load_weigth_for_model(model=LMSoftmax_weight,pretrained_path=LMSOFTMAX_WEIGHT_PTH,device=device)
LMSoftmax_weight.eval()
CNN_classfier = Filter_ID_predictor_from_1DCNN_LMSoftmax(MODEL_PATH, MATFILE_PATH, fs, LMSoftmax_weight, device)
valid_data = MyNoiseDataset(VALIDATTION_FILE,sheet)
valid_dataloader = create_data_loader(valid_data,BATCH_SIZE)
_, average_acc = tst_accuracy_of_model(valid_dataloader,CNN_classfier)
print(f"The average accuracy is {average_acc}")
#-----------------------------------------------------------------
if __name__ == "__main__":
if torch.cuda.is_available():
device = "cuda"
else:
device = "cpu"
print(f"Using {device}")
#device = "cpu"
MATFILE_PATH = 'Pre-train Control filter.mat'
#----
FILE_NAME_PATH = "Bandlimited_filter.mat"
FILE_NAME_PATH = "DesignBand_filter_v1.mat"
MATFILE_PATH = FILE_NAME_PATH
#-----
MODEL_PATH = "feedforwardnet.pth"
# MODEL_PATH = "feedforwardnet_Nway_Finetuned.pth"
#MODEL_PATH = 'feedforwardnet_LMSoftMax.pth'
VALIDATTION_FILE = "testing_data"
VALIDATTION_FILE = "testing_data_v1"
sheet = "Index.csv"
BATCH_SIZE = 100
fs = 16000
CNN_classfier = Filter_ID_predictor(MODEL_PATH, MATFILE_PATH, fs, device)
valid_data = MyNoiseDataset(VALIDATTION_FILE,sheet)
valid_dataloader = create_data_loader(valid_data,BATCH_SIZE)
_, average_acc = tst_accuracy_of_model(valid_dataloader,CNN_classfier)
print(f"The average accuracy is {average_acc}")