|
|
--- |
|
|
license: cc-by-4.0 |
|
|
metrics: |
|
|
- accuracy |
|
|
pipeline_tag: image-classification |
|
|
tags: |
|
|
- arXiv:1409.1556 |
|
|
datasets: |
|
|
- ILSVRC/imagenet-1k |
|
|
--- |
|
|
|
|
|
# VGG16 |
|
|
|
|
|
VGG16 model from ONNX Model Zoo |
|
|
|
|
|
## Aidge support |
|
|
|
|
|
> Note: We tested this network for the following features. If you encounter any error please open an [issue](https://gitlab.eclipse.org/groups/eclipse/aidge/-/issues). Features not tested in CI may not be functional. |
|
|
|
|
|
| Feature | Tested | Tested in CI | |
|
|
| :---------: | :----------: | :-----: | |
|
|
| ONNX import | ✔ | ✔ | |
|
|
| Backend CPU | ✔ | ✔ | |
|
|
| Export CPP | ✔ | ❌ | |
|
|
|
|
|
|
|
|
## ImageNet 224 |
|
|
|
|
|
- Origin: https://github.com/onnx/models/tree/main/validated/vision/classification/vgg |
|
|
- Input: |
|
|
- size: \[N, 3, 224, 224\] |
|
|
- format: NCHW |
|
|
- Output: |
|
|
- size: [1000] |
|
|
- ONNX: |
|
|
- opset: 18 |
|
|
- Operators: 38 (7 types) |
|
|
- FC: 1 |
|
|
- Flatten: 1 |
|
|
- GlobalAveragePooling : 1 |
|
|
- ReLU : 15 |
|
|
- Conv2D : 2 |
|
|
- MaxPooling2D : 5 |
|
|
- PaddedConv2D : 13 |
