Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
This set contains two models trained in the context of a thesis on emotion recognition in educational settings, using deep learning techniques.
|
| 2 |
|
| 3 |
- **FacialModel-8177.h5**: This model was developed as an ensemble combining convolutional neural networks, the VGG19 model, and Keras Tuner techniques. It was trained using two widely recognized datasets in the field of emotion recognition: FER2013 and CK+48. To address class imbalance, two complementary strategies were applied: (1) the use of weighted classes to adjust the importance of each class based on its representation, and (2) the generation of synthetic data through data augmentation techniques, which improved the distribution of emotions and the model’s generalization.
|
|
@@ -8,5 +31,4 @@ Both models are in `.h5` format and were developed using TensorFlow and Keras. T
|
|
| 8 |
|
| 9 |
This resource has been shared for scientific reproducibility and to support academic and technical analysis.
|
| 10 |
|
| 11 |
-
**Contact:** [email protected], [email protected]
|
| 12 |
-
|
|
|
|
| 1 |
+
---
|
| 2 |
+
tags:
|
| 3 |
+
- emotion-recognition
|
| 4 |
+
- education
|
| 5 |
+
- keras
|
| 6 |
+
- tensorflow
|
| 7 |
+
- multimodal
|
| 8 |
+
language:
|
| 9 |
+
- en
|
| 10 |
+
license: mit
|
| 11 |
+
datasets:
|
| 12 |
+
- fer2013
|
| 13 |
+
- ckplus
|
| 14 |
+
- isear
|
| 15 |
+
- meld
|
| 16 |
+
library_name: keras
|
| 17 |
+
metrics:
|
| 18 |
+
- accuracy
|
| 19 |
+
- precision
|
| 20 |
+
- recall
|
| 21 |
+
- f1
|
| 22 |
+
- perplexity
|
| 23 |
+
---
|
| 24 |
This set contains two models trained in the context of a thesis on emotion recognition in educational settings, using deep learning techniques.
|
| 25 |
|
| 26 |
- **FacialModel-8177.h5**: This model was developed as an ensemble combining convolutional neural networks, the VGG19 model, and Keras Tuner techniques. It was trained using two widely recognized datasets in the field of emotion recognition: FER2013 and CK+48. To address class imbalance, two complementary strategies were applied: (1) the use of weighted classes to adjust the importance of each class based on its representation, and (2) the generation of synthetic data through data augmentation techniques, which improved the distribution of emotions and the model’s generalization.
|
|
|
|
| 31 |
|
| 32 |
This resource has been shared for scientific reproducibility and to support academic and technical analysis.
|
| 33 |
|
| 34 |
+
**Contact:** [email protected], [email protected]
|
|
|