-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathmodel.py
137 lines (119 loc) · 4.9 KB
/
model.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
from torch import nn
import torch.nn.functional as F
from torch.nn.utils.rnn import pack_padded_sequence, pad_packed_sequence
class PoseS1(nn.Module):
def __init__(self):
super(PoseS1, self).__init__()
"""
PoseS1
输入归一化基于根坐标系的加速度和旋转矩阵
输入除关节点外5个关节点的坐标，以根坐标系为准
"""
self.L1 = nn.Dropout(p=0.2)
self.L2 = nn.Linear(72, 256) # relu
self.L3 = nn.LSTM(256, 256, bidirectional=True, batch_first=True, num_layers=2)
#self.L4 = nn.LSTM(512, 256, bidirectional=True, batch_first=True)
self.L5= nn.Linear(512, 15)
def forward(self, inputs, seq_lens):
total_length = inputs.shape[1]
output = self.L1(inputs)
output = F.relu(self.L2(output))
x_packed = pack_padded_sequence(output, seq_lens, batch_first=True, enforce_sorted=False)
output, _ = self.L3(x_packed)
#output, _ = self.L4(output)
output, length = pad_packed_sequence(output, batch_first=True, total_length=total_length)
output = self.L5(output)
return output
class PoseS2(nn.Module):
def __init__(self):
super(PoseS2, self).__init__()
"""
PoseS2
输入x1
输入除关节点外23个关节点的坐标，以根坐标系为准
"""
self.L1 = nn.Dropout(p=0.2)
self.L2 = nn.Linear(87, 64) # relu
self.L3 = nn.LSTM(64, 64, bidirectional=True, batch_first=True, num_layers=2)
#self.L4 = nn.LSTM(128, 64, bidirectional=True, batch_first=True)
self.L5= nn.Linear(128, 69)
def forward(self, inputs, seq_lens):
total_length = inputs.shape[1]
output = self.L1(inputs)
output = F.relu(self.L2(output))
x_packed = pack_padded_sequence(output, seq_lens, batch_first=True, enforce_sorted=False)
output, _ = self.L3(x_packed)
#output, _ = self.L4(output)
output, length = pad_packed_sequence(output, batch_first=True, total_length=total_length)
output = self.L5(output)
return output
class PoseS3(nn.Module):
def __init__(self):
super(PoseS3, self).__init__()
"""
PoseS3
输入x2
输入除关节点外23个关节点的旋转
"""
self.L1 = nn.Dropout(p=0.2)
self.L2 = nn.Linear(141, 128) # relu
self.L3 = nn.LSTM(128, 128, bidirectional=True, batch_first=True, num_layers=2)
#self.L4 = nn.LSTM(256, 128, bidirectional=True, batch_first=True)
self.L5= nn.Linear(256, 90)
def forward(self, inputs, seq_lens):
total_length = inputs.shape[1]
output = self.L1(inputs)
output = F.relu(self.L2(output))
x_packed = pack_padded_sequence(output, seq_lens, batch_first=True, enforce_sorted=False)
output, _ = self.L3(x_packed)
#output, _ = self.L4(output)
output, length = pad_packed_sequence(output, batch_first=True, total_length=total_length)
output = self.L5(output)
return output
class TransB1(nn.Module):
def __init__(self):
super(TransB1, self).__init__()
"""
TransB1
输入x1
输入支撑脚的概率
"""
self.L1 = nn.Dropout(p=0.2)
self.L2 = nn.Linear(87, 64) # relu
self.L3 = nn.LSTM(64, 64, bidirectional=True, batch_first=True, num_layers=2)
#self.L4 = nn.LSTM(128, 64, bidirectional=True, batch_first=True)
self.L5= nn.Linear(128, 2)
self.af = nn.Sigmoid()
def forward(self, inputs, seq_lens):
total_length = inputs.shape[1]
output = self.L1(inputs)
output = F.relu(self.L2(output))
x_packed = pack_padded_sequence(output, seq_lens, batch_first=True, enforce_sorted=False)
output, _ = self.L3(x_packed)
#output, _ = self.L4(output)
output, length = pad_packed_sequence(output, batch_first=True, total_length=total_length)
output = self.af(self.L5(output))
return output
class TransB2(nn.Module):
def __init__(self):
super(TransB2, self).__init__()
"""
TransB2
输入x2
输入根位移
"""
self.L1 = nn.Dropout(p=0.2)
self.L2 = nn.Linear(141, 256) # relu
self.L3 = nn.LSTM(256, 256, bidirectional=False, batch_first=True, num_layers=2)
#self.L4 = nn.LSTM(256, 256, bidirectional=False, batch_first=True)
self.L5= nn.Linear(256, 3)
def forward(self, inputs, seq_lens):
total_length = inputs.shape[1]
output = self.L1(inputs)
output = F.relu(self.L2(output))
x_packed = pack_padded_sequence(output, seq_lens, batch_first=True, enforce_sorted=False)
output, _ = self.L3(x_packed)
#output, _ = self.L4(output)
output, length = pad_packed_sequence(output, batch_first=True, total_length=total_length)
output = self.L5(output)
return output