Comments (3)
nicovandenhooff
commented on June 14, 2024
AlexNet
- PyTorch source code
- Model summary:
==========================================================================================
Layer (type:depth-idx) Output Shape Param #
==========================================================================================
AlexNet -- --
├─Sequential: 1-1 [64, 256, 3, 3] --
│ └─Conv2d: 2-1 [64, 64, 31, 31] 23,296
│ └─ReLU: 2-2 [64, 64, 31, 31] --
│ └─MaxPool2d: 2-3 [64, 64, 15, 15] --
│ └─Conv2d: 2-4 [64, 192, 15, 15] 307,392
│ └─ReLU: 2-5 [64, 192, 15, 15] --
│ └─MaxPool2d: 2-6 [64, 192, 7, 7] --
│ └─Conv2d: 2-7 [64, 384, 7, 7] 663,936
│ └─ReLU: 2-8 [64, 384, 7, 7] --
│ └─Conv2d: 2-9 [64, 256, 7, 7] 884,992
│ └─ReLU: 2-10 [64, 256, 7, 7] --
│ └─Conv2d: 2-11 [64, 256, 7, 7] 590,080
│ └─ReLU: 2-12 [64, 256, 7, 7] --
│ └─MaxPool2d: 2-13 [64, 256, 3, 3] --
├─AdaptiveAvgPool2d: 1-2 [64, 256, 6, 6] --
├─Sequential: 1-3 [64, 1000] --
│ └─Dropout: 2-14 [64, 9216] --
│ └─Linear: 2-15 [64, 4096] 37,752,832
│ └─ReLU: 2-16 [64, 4096] --
│ └─Dropout: 2-17 [64, 4096] --
│ └─Linear: 2-18 [64, 4096] 16,781,312
│ └─ReLU: 2-19 [64, 4096] --
│ └─Linear: 2-20 [64, 1000] 4,097,000
==========================================================================================
Total params: 61,100,840
Trainable params: 61,100,840
Non-trainable params: 0
Total mult-adds (G): 16.32
==========================================================================================
Input size (MB): 12.58
Forward/backward pass size (MB): 80.79
Params size (MB): 244.40
Estimated Total Size (MB): 337.78
==========================================================================================
- Last layer:
self.classifier = nn.Sequential(
nn.Dropout(p=dropout),
nn.Linear(256 * 6 * 6, 4096),
nn.ReLU(inplace=True),
nn.Dropout(p=dropout),
nn.Linear(4096, 4096),
nn.ReLU(inplace=True),
nn.Linear(4096, num_classes),
)
from indoor-scene-detector.
nicovandenhooff
commented on June 14, 2024
DenseNet
- Note: Options include
densenet121,densenet161,densenet169,densenet201 - PyTorch source code
- Model summary for
densenet121:
==========================================================================================
Layer (type:depth-idx) Output Shape Param #
==========================================================================================
DenseNet -- --
├─Sequential: 1-1 [64, 1024, 4, 4] --
│ └─Conv2d: 2-1 [64, 64, 64, 64] 9,408
│ └─BatchNorm2d: 2-2 [64, 64, 64, 64] 128
│ └─ReLU: 2-3 [64, 64, 64, 64] --
│ └─MaxPool2d: 2-4 [64, 64, 32, 32] --
│ └─_DenseBlock: 2-5 [64, 256, 32, 32] --
│ │ └─_DenseLayer: 3-1 [64, 32, 32, 32] 45,440
│ │ └─_DenseLayer: 3-2 [64, 32, 32, 32] 49,600
│ │ └─_DenseLayer: 3-3 [64, 32, 32, 32] 53,760
│ │ └─_DenseLayer: 3-4 [64, 32, 32, 32] 57,920
│ │ └─_DenseLayer: 3-5 [64, 32, 32, 32] 62,080
│ │ └─_DenseLayer: 3-6 [64, 32, 32, 32] 66,240
│ └─_Transition: 2-6 [64, 128, 16, 16] --
│ │ └─BatchNorm2d: 3-7 [64, 256, 32, 32] 512
│ │ └─ReLU: 3-8 [64, 256, 32, 32] --
│ │ └─Conv2d: 3-9 [64, 128, 32, 32] 32,768
│ │ └─AvgPool2d: 3-10 [64, 128, 16, 16] --
│ └─_DenseBlock: 2-7 [64, 512, 16, 16] --
│ │ └─_DenseLayer: 3-11 [64, 32, 16, 16] 53,760
│ │ └─_DenseLayer: 3-12 [64, 32, 16, 16] 57,920
│ │ └─_DenseLayer: 3-13 [64, 32, 16, 16] 62,080
│ │ └─_DenseLayer: 3-14 [64, 32, 16, 16] 66,240
│ │ └─_DenseLayer: 3-15 [64, 32, 16, 16] 70,400
│ │ └─_DenseLayer: 3-16 [64, 32, 16, 16] 74,560
│ │ └─_DenseLayer: 3-17 [64, 32, 16, 16] 78,720
│ │ └─_DenseLayer: 3-18 [64, 32, 16, 16] 82,880
│ │ └─_DenseLayer: 3-19 [64, 32, 16, 16] 87,040
│ │ └─_DenseLayer: 3-20 [64, 32, 16, 16] 91,200
│ │ └─_DenseLayer: 3-21 [64, 32, 16, 16] 95,360
│ │ └─_DenseLayer: 3-22 [64, 32, 16, 16] 99,520
│ └─_Transition: 2-8 [64, 256, 8, 8] --
│ │ └─BatchNorm2d: 3-23 [64, 512, 16, 16] 1,024
│ │ └─ReLU: 3-24 [64, 512, 16, 16] --
│ │ └─Conv2d: 3-25 [64, 256, 16, 16] 131,072
│ │ └─AvgPool2d: 3-26 [64, 256, 8, 8] --
│ └─_DenseBlock: 2-9 [64, 1024, 8, 8] --
│ │ └─_DenseLayer: 3-27 [64, 32, 8, 8] 70,400
│ │ └─_DenseLayer: 3-28 [64, 32, 8, 8] 74,560
│ │ └─_DenseLayer: 3-29 [64, 32, 8, 8] 78,720
│ │ └─_DenseLayer: 3-30 [64, 32, 8, 8] 82,880
│ │ └─_DenseLayer: 3-31 [64, 32, 8, 8] 87,040
│ │ └─_DenseLayer: 3-32 [64, 32, 8, 8] 91,200
│ │ └─_DenseLayer: 3-33 [64, 32, 8, 8] 95,360
│ │ └─_DenseLayer: 3-34 [64, 32, 8, 8] 99,520
│ │ └─_DenseLayer: 3-35 [64, 32, 8, 8] 103,680
│ │ └─_DenseLayer: 3-36 [64, 32, 8, 8] 107,840
│ │ └─_DenseLayer: 3-37 [64, 32, 8, 8] 112,000
│ │ └─_DenseLayer: 3-38 [64, 32, 8, 8] 116,160
│ │ └─_DenseLayer: 3-39 [64, 32, 8, 8] 120,320
│ │ └─_DenseLayer: 3-40 [64, 32, 8, 8] 124,480
│ │ └─_DenseLayer: 3-41 [64, 32, 8, 8] 128,640
│ │ └─_DenseLayer: 3-42 [64, 32, 8, 8] 132,800
│ │ └─_DenseLayer: 3-43 [64, 32, 8, 8] 136,960
│ │ └─_DenseLayer: 3-44 [64, 32, 8, 8] 141,120
│ │ └─_DenseLayer: 3-45 [64, 32, 8, 8] 145,280
│ │ └─_DenseLayer: 3-46 [64, 32, 8, 8] 149,440
│ │ └─_DenseLayer: 3-47 [64, 32, 8, 8] 153,600
│ │ └─_DenseLayer: 3-48 [64, 32, 8, 8] 157,760
│ │ └─_DenseLayer: 3-49 [64, 32, 8, 8] 161,920
│ │ └─_DenseLayer: 3-50 [64, 32, 8, 8] 166,080
│ └─_Transition: 2-10 [64, 512, 4, 4] --
│ │ └─BatchNorm2d: 3-51 [64, 1024, 8, 8] 2,048
│ │ └─ReLU: 3-52 [64, 1024, 8, 8] --
│ │ └─Conv2d: 3-53 [64, 512, 8, 8] 524,288
│ │ └─AvgPool2d: 3-54 [64, 512, 4, 4] --
│ └─_DenseBlock: 2-11 [64, 1024, 4, 4] --
│ │ └─_DenseLayer: 3-55 [64, 32, 4, 4] 103,680
│ │ └─_DenseLayer: 3-56 [64, 32, 4, 4] 107,840
│ │ └─_DenseLayer: 3-57 [64, 32, 4, 4] 112,000
│ │ └─_DenseLayer: 3-58 [64, 32, 4, 4] 116,160
│ │ └─_DenseLayer: 3-59 [64, 32, 4, 4] 120,320
│ │ └─_DenseLayer: 3-60 [64, 32, 4, 4] 124,480
│ │ └─_DenseLayer: 3-61 [64, 32, 4, 4] 128,640
│ │ └─_DenseLayer: 3-62 [64, 32, 4, 4] 132,800
│ │ └─_DenseLayer: 3-63 [64, 32, 4, 4] 136,960
│ │ └─_DenseLayer: 3-64 [64, 32, 4, 4] 141,120
│ │ └─_DenseLayer: 3-65 [64, 32, 4, 4] 145,280
│ │ └─_DenseLayer: 3-66 [64, 32, 4, 4] 149,440
│ │ └─_DenseLayer: 3-67 [64, 32, 4, 4] 153,600
│ │ └─_DenseLayer: 3-68 [64, 32, 4, 4] 157,760
│ │ └─_DenseLayer: 3-69 [64, 32, 4, 4] 161,920
│ │ └─_DenseLayer: 3-70 [64, 32, 4, 4] 166,080
│ └─BatchNorm2d: 2-12 [64, 1024, 4, 4] 2,048
├─Linear: 1-2 [64, 1000] 1,025,000
==========================================================================================
Total params: 7,978,856
Trainable params: 7,978,856
Non-trainable params: 0
Total mult-adds (G): 59.28
==========================================================================================
Input size (MB): 12.58
Forward/backward pass size (MB): 3773.29
Params size (MB): 31.92
Estimated Total Size (MB): 3817.79
==========================================================================================
- Last layer:
self.classifier = nn.Linear(num_features, num_classes)
from indoor-scene-detector.
nicovandenhooff
commented on June 14, 2024
ResNet
- Note: options include
resnet101,resnet152,resnet101,resnet18,resnet50 - PyTorch source code
- Model summary for
resnet18
==========================================================================================
Layer (type:depth-idx) Output Shape Param #
==========================================================================================
ResNet -- --
├─Conv2d: 1-1 [64, 64, 64, 64] 9,408
├─BatchNorm2d: 1-2 [64, 64, 64, 64] 128
├─ReLU: 1-3 [64, 64, 64, 64] --
├─MaxPool2d: 1-4 [64, 64, 32, 32] --
├─Sequential: 1-5 [64, 64, 32, 32] --
│ └─BasicBlock: 2-1 [64, 64, 32, 32] --
│ │ └─Conv2d: 3-1 [64, 64, 32, 32] 36,864
│ │ └─BatchNorm2d: 3-2 [64, 64, 32, 32] 128
│ │ └─ReLU: 3-3 [64, 64, 32, 32] --
│ │ └─Conv2d: 3-4 [64, 64, 32, 32] 36,864
│ │ └─BatchNorm2d: 3-5 [64, 64, 32, 32] 128
│ │ └─ReLU: 3-6 [64, 64, 32, 32] --
│ └─BasicBlock: 2-2 [64, 64, 32, 32] --
│ │ └─Conv2d: 3-7 [64, 64, 32, 32] 36,864
│ │ └─BatchNorm2d: 3-8 [64, 64, 32, 32] 128
│ │ └─ReLU: 3-9 [64, 64, 32, 32] --
│ │ └─Conv2d: 3-10 [64, 64, 32, 32] 36,864
│ │ └─BatchNorm2d: 3-11 [64, 64, 32, 32] 128
│ │ └─ReLU: 3-12 [64, 64, 32, 32] --
├─Sequential: 1-6 [64, 128, 16, 16] --
│ └─BasicBlock: 2-3 [64, 128, 16, 16] --
│ │ └─Conv2d: 3-13 [64, 128, 16, 16] 73,728
│ │ └─BatchNorm2d: 3-14 [64, 128, 16, 16] 256
│ │ └─ReLU: 3-15 [64, 128, 16, 16] --
│ │ └─Conv2d: 3-16 [64, 128, 16, 16] 147,456
│ │ └─BatchNorm2d: 3-17 [64, 128, 16, 16] 256
│ │ └─Sequential: 3-18 [64, 128, 16, 16] 8,448
│ │ └─ReLU: 3-19 [64, 128, 16, 16] --
│ └─BasicBlock: 2-4 [64, 128, 16, 16] --
│ │ └─Conv2d: 3-20 [64, 128, 16, 16] 147,456
│ │ └─BatchNorm2d: 3-21 [64, 128, 16, 16] 256
│ │ └─ReLU: 3-22 [64, 128, 16, 16] --
│ │ └─Conv2d: 3-23 [64, 128, 16, 16] 147,456
│ │ └─BatchNorm2d: 3-24 [64, 128, 16, 16] 256
│ │ └─ReLU: 3-25 [64, 128, 16, 16] --
├─Sequential: 1-7 [64, 256, 8, 8] --
│ └─BasicBlock: 2-5 [64, 256, 8, 8] --
│ │ └─Conv2d: 3-26 [64, 256, 8, 8] 294,912
│ │ └─BatchNorm2d: 3-27 [64, 256, 8, 8] 512
│ │ └─ReLU: 3-28 [64, 256, 8, 8] --
│ │ └─Conv2d: 3-29 [64, 256, 8, 8] 589,824
│ │ └─BatchNorm2d: 3-30 [64, 256, 8, 8] 512
│ │ └─Sequential: 3-31 [64, 256, 8, 8] 33,280
│ │ └─ReLU: 3-32 [64, 256, 8, 8] --
│ └─BasicBlock: 2-6 [64, 256, 8, 8] --
│ │ └─Conv2d: 3-33 [64, 256, 8, 8] 589,824
│ │ └─BatchNorm2d: 3-34 [64, 256, 8, 8] 512
│ │ └─ReLU: 3-35 [64, 256, 8, 8] --
│ │ └─Conv2d: 3-36 [64, 256, 8, 8] 589,824
│ │ └─BatchNorm2d: 3-37 [64, 256, 8, 8] 512
│ │ └─ReLU: 3-38 [64, 256, 8, 8] --
├─Sequential: 1-8 [64, 512, 4, 4] --
│ └─BasicBlock: 2-7 [64, 512, 4, 4] --
│ │ └─Conv2d: 3-39 [64, 512, 4, 4] 1,179,648
│ │ └─BatchNorm2d: 3-40 [64, 512, 4, 4] 1,024
│ │ └─ReLU: 3-41 [64, 512, 4, 4] --
│ │ └─Conv2d: 3-42 [64, 512, 4, 4] 2,359,296
│ │ └─BatchNorm2d: 3-43 [64, 512, 4, 4] 1,024
│ │ └─Sequential: 3-44 [64, 512, 4, 4] 132,096
│ │ └─ReLU: 3-45 [64, 512, 4, 4] --
│ └─BasicBlock: 2-8 [64, 512, 4, 4] --
│ │ └─Conv2d: 3-46 [64, 512, 4, 4] 2,359,296
│ │ └─BatchNorm2d: 3-47 [64, 512, 4, 4] 1,024
│ │ └─ReLU: 3-48 [64, 512, 4, 4] --
│ │ └─Conv2d: 3-49 [64, 512, 4, 4] 2,359,296
│ │ └─BatchNorm2d: 3-50 [64, 512, 4, 4] 1,024
│ │ └─ReLU: 3-51 [64, 512, 4, 4] --
├─AdaptiveAvgPool2d: 1-9 [64, 512, 1, 1] --
├─Linear: 1-10 [64, 1000] 513,000
==========================================================================================
Total params: 11,689,512
Trainable params: 11,689,512
Non-trainable params: 0
Total mult-adds (G): 37.93
==========================================================================================
Input size (MB): 12.58
Forward/backward pass size (MB): 830.98
Params size (MB): 46.76
Estimated Total Size (MB): 890.33
==========================================================================================
- Last layer:
self.fc = nn.Linear(512 * block.expansion, num_classes)
from indoor-scene-detector.
Related Issues (15)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from indoor-scene-detector.