This is a PyTorch implementation of Cocktail Causal Container proposed by our paper "End to End Video based Cocktail Causal Container for Blood Pressure Estimation and Glucose Prediction".
| Model | acc_bp@1 | acc_bg@1 | Model |
|---|---|---|---|
| Cocktail Causal Container | 75.0 | 91.7 | soon |
| Model | overall rmse_bg | acc_bp@1 | acc_bg@1 | Model |
|---|---|---|---|---|
| Cocktail Causal Container | 0.766 | 89.0 | 89.0 | soon |
The code and model will be released soon.
Layer (type) Output Shape Param #
================================================================
Conv2d-1 [-1, 32, 50, 50] 5,184
SiLU-2 [-1, 32, 50, 50] 0
BatchNormAct2d-3 [-1, 32, 50, 50] 64
ConvBnAct-4 [-1, 32, 50, 50] 0
Conv2d-5 [-1, 32, 50, 50] 288
Identity-6 [-1, 32, 50, 50] 0
BatchNormAct2d-7 [-1, 32, 50, 50] 64
ConvBnAct-8 [-1, 32, 50, 50] 0
ReLU6-9 [-1, 32, 50, 50] 0
Conv2d-10 [-1, 16, 50, 50] 512
Identity-11 [-1, 16, 50, 50] 0
BatchNormAct2d-12 [-1, 16, 50, 50] 32
ConvBnAct-13 [-1, 16, 50, 50] 0
LinearBottleneck-14 [-1, 16, 50, 50] 0
Conv2d-15 [-1, 96, 50, 50] 1,536
SiLU-16 [-1, 96, 50, 50] 0
BatchNormAct2d-17 [-1, 96, 50, 50] 192
ConvBnAct-18 [-1, 96, 50, 50] 0
Conv2d-19 [-1, 96, 25, 25] 864
Identity-20 [-1, 96, 25, 25] 0
BatchNormAct2d-21 [-1, 96, 25, 25] 192
ConvBnAct-22 [-1, 96, 25, 25] 0
ReLU6-23 [-1, 96, 25, 25] 0
Conv2d-24 [-1, 27, 25, 25] 2,592
Identity-25 [-1, 27, 25, 25] 0
BatchNormAct2d-26 [-1, 27, 25, 25] 54
ConvBnAct-27 [-1, 27, 25, 25] 0
LinearBottleneck-28 [-1, 27, 25, 25] 0
Conv2d-29 [-1, 162, 25, 25] 4,374
SiLU-30 [-1, 162, 25, 25] 0
BatchNormAct2d-31 [-1, 162, 25, 25] 324
ConvBnAct-32 [-1, 162, 25, 25] 0
Conv2d-33 [-1, 162, 25, 25] 1,458
Identity-34 [-1, 162, 25, 25] 0
BatchNormAct2d-35 [-1, 162, 25, 25] 324
ConvBnAct-36 [-1, 162, 25, 25] 0
ReLU6-37 [-1, 162, 25, 25] 0
Conv2d-38 [-1, 38, 25, 25] 6,156
Identity-39 [-1, 38, 25, 25] 0
BatchNormAct2d-40 [-1, 38, 25, 25] 76
ConvBnAct-41 [-1, 38, 25, 25] 0
LinearBottleneck-42 [-1, 38, 25, 25] 0
Conv2d-43 [-1, 228, 25, 25] 8,664
SiLU-44 [-1, 228, 25, 25] 0
BatchNormAct2d-45 [-1, 228, 25, 25] 456
ConvBnAct-46 [-1, 228, 25, 25] 0
Conv2d-47 [-1, 228, 13, 13] 2,052
Identity-48 [-1, 228, 13, 13] 0
BatchNormAct2d-49 [-1, 228, 13, 13] 456
ConvBnAct-50 [-1, 228, 13, 13] 0
Conv2d-51 [-1, 19, 1, 1] 4,351
BatchNorm2d-52 [-1, 19, 1, 1] 38
ReLU-53 [-1, 19, 1, 1] 0
Conv2d-54 [-1, 228, 1, 1] 4,560
Sigmoid-55 [-1, 228, 1, 1] 0
SEWithNorm-56 [-1, 228, 13, 13] 0
ReLU6-57 [-1, 228, 13, 13] 0
Conv2d-58 [-1, 50, 13, 13] 11,400
Identity-59 [-1, 50, 13, 13] 0
BatchNormAct2d-60 [-1, 50, 13, 13] 100
ConvBnAct-61 [-1, 50, 13, 13] 0
LinearBottleneck-62 [-1, 50, 13, 13] 0
Conv2d-63 [-1, 300, 13, 13] 15,000
SiLU-64 [-1, 300, 13, 13] 0
BatchNormAct2d-65 [-1, 300, 13, 13] 600
ConvBnAct-66 [-1, 300, 13, 13] 0
Conv2d-67 [-1, 300, 13, 13] 2,700
Identity-68 [-1, 300, 13, 13] 0
BatchNormAct2d-69 [-1, 300, 13, 13] 600
ConvBnAct-70 [-1, 300, 13, 13] 0
Conv2d-71 [-1, 25, 1, 1] 7,525
BatchNorm2d-72 [-1, 25, 1, 1] 50
ReLU-73 [-1, 25, 1, 1] 0
Conv2d-74 [-1, 300, 1, 1] 7,800
Sigmoid-75 [-1, 300, 1, 1] 0
SEWithNorm-76 [-1, 300, 13, 13] 0
ReLU6-77 [-1, 300, 13, 13] 0
Conv2d-78 [-1, 61, 13, 13] 18,300
Identity-79 [-1, 61, 13, 13] 0
BatchNormAct2d-80 [-1, 61, 13, 13] 122
ConvBnAct-81 [-1, 61, 13, 13] 0
LinearBottleneck-82 [-1, 61, 13, 13] 0
Conv2d-83 [-1, 366, 13, 13] 22,326
SiLU-84 [-1, 366, 13, 13] 0
BatchNormAct2d-85 [-1, 366, 13, 13] 732
ConvBnAct-86 [-1, 366, 13, 13] 0
Conv2d-87 [-1, 366, 7, 7] 3,294
Identity-88 [-1, 366, 7, 7] 0
BatchNormAct2d-89 [-1, 366, 7, 7] 732
ConvBnAct-90 [-1, 366, 7, 7] 0
Conv2d-91 [-1, 30, 1, 1] 11,010
BatchNorm2d-92 [-1, 30, 1, 1] 60
ReLU-93 [-1, 30, 1, 1] 0
Conv2d-94 [-1, 366, 1, 1] 11,346
Sigmoid-95 [-1, 366, 1, 1] 0
SEWithNorm-96 [-1, 366, 7, 7] 0
ReLU6-97 [-1, 366, 7, 7] 0
Conv2d-98 [-1, 72, 7, 7] 26,352
Identity-99 [-1, 72, 7, 7] 0
BatchNormAct2d-100 [-1, 72, 7, 7] 144
ConvBnAct-101 [-1, 72, 7, 7] 0
LinearBottleneck-102 [-1, 72, 7, 7] 0
Conv2d-103 [-1, 432, 7, 7] 31,104
SiLU-104 [-1, 432, 7, 7] 0
BatchNormAct2d-105 [-1, 432, 7, 7] 864
ConvBnAct-106 [-1, 432, 7, 7] 0
Conv2d-107 [-1, 432, 7, 7] 3,888
Identity-108 [-1, 432, 7, 7] 0
BatchNormAct2d-109 [-1, 432, 7, 7] 864
ConvBnAct-110 [-1, 432, 7, 7] 0
Conv2d-111 [-1, 36, 1, 1] 15,588
BatchNorm2d-112 [-1, 36, 1, 1] 72
ReLU-113 [-1, 36, 1, 1] 0
Conv2d-114 [-1, 432, 1, 1] 15,984
Sigmoid-115 [-1, 432, 1, 1] 0
SEWithNorm-116 [-1, 432, 7, 7] 0
ReLU6-117 [-1, 432, 7, 7] 0
Conv2d-118 [-1, 84, 7, 7] 36,288
Identity-119 [-1, 84, 7, 7] 0
BatchNormAct2d-120 [-1, 84, 7, 7] 168
ConvBnAct-121 [-1, 84, 7, 7] 0
LinearBottleneck-122 [-1, 84, 7, 7] 0
Conv2d-123 [-1, 504, 7, 7] 42,336
SiLU-124 [-1, 504, 7, 7] 0
BatchNormAct2d-125 [-1, 504, 7, 7] 1,008
ConvBnAct-126 [-1, 504, 7, 7] 0
Conv2d-127 [-1, 504, 7, 7] 4,536
Identity-128 [-1, 504, 7, 7] 0
BatchNormAct2d-129 [-1, 504, 7, 7] 1,008
ConvBnAct-130 [-1, 504, 7, 7] 0
Conv2d-131 [-1, 42, 1, 1] 21,210
BatchNorm2d-132 [-1, 42, 1, 1] 84
ReLU-133 [-1, 42, 1, 1] 0
Conv2d-134 [-1, 504, 1, 1] 21,672
Sigmoid-135 [-1, 504, 1, 1] 0
SEWithNorm-136 [-1, 504, 7, 7] 0
ReLU6-137 [-1, 504, 7, 7] 0
Conv2d-138 [-1, 95, 7, 7] 47,880
Identity-139 [-1, 95, 7, 7] 0
BatchNormAct2d-140 [-1, 95, 7, 7] 190
ConvBnAct-141 [-1, 95, 7, 7] 0
LinearBottleneck-142 [-1, 95, 7, 7] 0
Conv2d-143 [-1, 570, 7, 7] 54,150
SiLU-144 [-1, 570, 7, 7] 0
BatchNormAct2d-145 [-1, 570, 7, 7] 1,140
ConvBnAct-146 [-1, 570, 7, 7] 0
Conv2d-147 [-1, 570, 7, 7] 5,130
Identity-148 [-1, 570, 7, 7] 0
BatchNormAct2d-149 [-1, 570, 7, 7] 1,140
ConvBnAct-150 [-1, 570, 7, 7] 0
Conv2d-151 [-1, 47, 1, 1] 26,837
BatchNorm2d-152 [-1, 47, 1, 1] 94
ReLU-153 [-1, 47, 1, 1] 0
Conv2d-154 [-1, 570, 1, 1] 27,360
Sigmoid-155 [-1, 570, 1, 1] 0
SEWithNorm-156 [-1, 570, 7, 7] 0
ReLU6-157 [-1, 570, 7, 7] 0
Conv2d-158 [-1, 106, 7, 7] 60,420
Identity-159 [-1, 106, 7, 7] 0
BatchNormAct2d-160 [-1, 106, 7, 7] 212
ConvBnAct-161 [-1, 106, 7, 7] 0
LinearBottleneck-162 [-1, 106, 7, 7] 0
Conv2d-163 [-1, 636, 7, 7] 67,416
SiLU-164 [-1, 636, 7, 7] 0
BatchNormAct2d-165 [-1, 636, 7, 7] 1,272
ConvBnAct-166 [-1, 636, 7, 7] 0
Conv2d-167 [-1, 636, 7, 7] 5,724
Identity-168 [-1, 636, 7, 7] 0
BatchNormAct2d-169 [-1, 636, 7, 7] 1,272
ConvBnAct-170 [-1, 636, 7, 7] 0
Conv2d-171 [-1, 53, 1, 1] 33,761
BatchNorm2d-172 [-1, 53, 1, 1] 106
ReLU-173 [-1, 53, 1, 1] 0
Conv2d-174 [-1, 636, 1, 1] 34,344
Sigmoid-175 [-1, 636, 1, 1] 0
SEWithNorm-176 [-1, 636, 7, 7] 0
ReLU6-177 [-1, 636, 7, 7] 0
Conv2d-178 [-1, 117, 7, 7] 74,412
Identity-179 [-1, 117, 7, 7] 0
BatchNormAct2d-180 [-1, 117, 7, 7] 234
ConvBnAct-181 [-1, 117, 7, 7] 0
LinearBottleneck-182 [-1, 117, 7, 7] 0
Conv2d-183 [-1, 702, 7, 7] 82,134
SiLU-184 [-1, 702, 7, 7] 0
BatchNormAct2d-185 [-1, 702, 7, 7] 1,404
ConvBnAct-186 [-1, 702, 7, 7] 0
Conv2d-187 [-1, 702, 7, 7] 6,318
Identity-188 [-1, 702, 7, 7] 0
BatchNormAct2d-189 [-1, 702, 7, 7] 1,404
ConvBnAct-190 [-1, 702, 7, 7] 0
Conv2d-191 [-1, 58, 1, 1] 40,774
BatchNorm2d-192 [-1, 58, 1, 1] 116
ReLU-193 [-1, 58, 1, 1] 0
Conv2d-194 [-1, 702, 1, 1] 41,418
Sigmoid-195 [-1, 702, 1, 1] 0
SEWithNorm-196 [-1, 702, 7, 7] 0
ReLU6-197 [-1, 702, 7, 7] 0
Conv2d-198 [-1, 128, 7, 7] 89,856
Identity-199 [-1, 128, 7, 7] 0
BatchNormAct2d-200 [-1, 128, 7, 7] 256
...
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent