Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- ylecun/mnist
|
| 4 |
+
metrics:
|
| 5 |
+
- accuracy
|
| 6 |
+
pipeline_tag: image-classification
|
| 7 |
+
library_name: torch
|
| 8 |
+
---
|
| 9 |
+
|
| 10 |
+
# LeNet
|
| 11 |
+
|
| 12 |
+
Toy model mainly used to showcase Aidge in various tutorial, for example: https://eclipse.dev/aidge/source/Tutorial/101_first_step.html
|
| 13 |
+
|
| 14 |
+
## Aidge support
|
| 15 |
+
|
| 16 |
+
> 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).
|
| 17 |
+
|
| 18 |
+
| Feature | Tested in CI |
|
| 19 |
+
| :---------: | :----------: |
|
| 20 |
+
| ONNX import | ✔ |
|
| 21 |
+
| Backend CPU | ❌ |
|
| 22 |
+
| Export CPP | ❌ |
|
| 23 |
+
|
| 24 |
+
## MNIST
|
| 25 |
+
|
| 26 |
+
* **Input**
|
| 27 |
+
* size: [N, 1, 28, 28]
|
| 28 |
+
* format: [N, C, H, W]
|
| 29 |
+
* preprocessing: `None`
|
| 30 |
+
* **Output**
|
| 31 |
+
* size: [N, 10]
|
| 32 |
+
|
| 33 |
+
### ONNX attributes
|
| 34 |
+
|
| 35 |
+
* Opset: <opset>
|
| 36 |
+
* Source: PyTorch
|
| 37 |
+
* Operators: 22 (6 types)
|
| 38 |
+
- Conv2D: 2
|
| 39 |
+
- FC: 3
|
| 40 |
+
- Flatten: 1
|
| 41 |
+
- MaxPooling2D: 2
|
| 42 |
+
- Producer: 10
|
| 43 |
+
- ReLU: 4
|
| 44 |
+
|
| 45 |
+
### Benchmark
|
| 46 |
+
|
| 47 |
+
> Coming soon
|