Update README.md
Browse files
README.md
CHANGED
|
@@ -61,7 +61,9 @@ For an image resolution of NxM and P classes and 0.25 alpha parameter :
|
|
| 61 |
# Performances
|
| 62 |
|
| 63 |
## Metrics
|
| 64 |
-
|
|
|
|
|
|
|
| 65 |
|
| 66 |
|
| 67 |
### Reference **NPU** memory footprint on food-101 dataset (see Accuracy for details on dataset)
|
|
|
|
| 61 |
# Performances
|
| 62 |
|
| 63 |
## Metrics
|
| 64 |
+
|
| 65 |
+
* Measures are done with default STM32Cube.AI configuration with enabled input / output allocated option.
|
| 66 |
+
* `tfs` stands for "transfer learning", meaning that the model backbone weights were initialized from a pre-trained model, then only the last layer was unfrozen during the training.
|
| 67 |
|
| 68 |
|
| 69 |
### Reference **NPU** memory footprint on food-101 dataset (see Accuracy for details on dataset)
|