Update Readme ST Model Zoo
Browse files
README.md
CHANGED
|
@@ -1,9 +1,3 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: other
|
| 3 |
-
license_name: sla0044
|
| 4 |
-
license_link: >-
|
| 5 |
-
https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/ST_pretrainedmodel_public_dataset/LICENSE.md
|
| 6 |
-
---
|
| 7 |
# IGN HAR model
|
| 8 |
|
| 9 |
## **Use case** : `Human activity recognition`
|
|
@@ -76,8 +70,8 @@ Dataset details: A custom dataset and not publically available, Number of classe
|
|
| 76 |
|
| 77 |
| Model | Format | Resolution | Accuracy (%)|
|
| 78 |
|:--------------------------------------------------------------------------------------------:|:------:|:----------:|:-----------:|
|
| 79 |
-
| [IGN wl 24](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/
|
| 80 |
-
| [IGN wl 48](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/
|
| 81 |
|
| 82 |
Confusion matrix for IGN wl 24 with Float32 weights for mobility_v1 dataset is given below.
|
| 83 |
|
|
@@ -91,8 +85,8 @@ Dataset details: [link](([WISDM]("https://www.cis.fordham.edu/wisdm/dataset.php"
|
|
| 91 |
|
| 92 |
| Model | Format | Resolution | Accuracy (%) |
|
| 93 |
|:-------------------------------------------------------------------------------------:|:-------:|:----------:|:-------------:|
|
| 94 |
-
| [IGN wl 24](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/
|
| 95 |
-
| [IGN wl 48](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/
|
| 96 |
|
| 97 |
|
| 98 |
## Retraining and Integration in a simple example:
|
|
@@ -107,5 +101,4 @@ Please refer to the stm32ai-modelzoo-services GitHub [here](https://github.com/S
|
|
| 107 |
“WISDM : Human activity recognition datasets". [Online]. Available: "https://www.cis.fordham.edu/wisdm/dataset.php".
|
| 108 |
|
| 109 |
<a id="2">[2]</a>
|
| 110 |
-
“Real-time human activity recognition from accelerometer data using Convolutional Neural Networks, Andrey Ignatove". [Online]. Available: "https://www.sciencedirect.com/science/article/abs/pii/S1568494617305665?via%3Dihub".
|
| 111 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# IGN HAR model
|
| 2 |
|
| 3 |
## **Use case** : `Human activity recognition`
|
|
|
|
| 70 |
|
| 71 |
| Model | Format | Resolution | Accuracy (%)|
|
| 72 |
|:--------------------------------------------------------------------------------------------:|:------:|:----------:|:-----------:|
|
| 73 |
+
| [IGN wl 24](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/ST_pretrainedmodel_public_dataset/mobility_v1/ign_wl_24/ign_wl_24.h5) | FLOAT32| 24 x 3 x 1 | 94.64 |
|
| 74 |
+
| [IGN wl 48](https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/ST_pretrainedmodel_public_dataset/mobility_v1/ign_wl_48/ign_wl_48.h5) | FLOAT32| 48 x 3 x 1 | 95.01 |
|
| 75 |
|
| 76 |
Confusion matrix for IGN wl 24 with Float32 weights for mobility_v1 dataset is given below.
|
| 77 |
|
|
|
|
| 85 |
|
| 86 |
| Model | Format | Resolution | Accuracy (%) |
|
| 87 |
|:-------------------------------------------------------------------------------------:|:-------:|:----------:|:-------------:|
|
| 88 |
+
| [IGN wl 24]((https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/ST_pretrainedmodel_public_datase/WISDM/ign_wl_24/ign_wl_24.h5) | FLOAT32 | 24 x 3 x 1 | 91.7 |
|
| 89 |
+
| [IGN wl 48]((https://github.com/STMicroelectronics/stm32ai-modelzoo/tree/main/human_activity_recognition/ign/ST_pretrainedmodel_public_datase/WISDM/ign_wl_48/ign_wl_48.h5) | FLOAT32 | 48 x 3 x 1 | 93.67 |
|
| 90 |
|
| 91 |
|
| 92 |
## Retraining and Integration in a simple example:
|
|
|
|
| 101 |
“WISDM : Human activity recognition datasets". [Online]. Available: "https://www.cis.fordham.edu/wisdm/dataset.php".
|
| 102 |
|
| 103 |
<a id="2">[2]</a>
|
| 104 |
+
“Real-time human activity recognition from accelerometer data using Convolutional Neural Networks, Andrey Ignatove". [Online]. Available: "https://www.sciencedirect.com/science/article/abs/pii/S1568494617305665?via%3Dihub".
|
|
|