DebertaV3ForAIS
Model Description
The model is based on the DeBERTa-v3 architecture, a transformer-based model that performs text classification tasks. It has been fine-tuned on a specific dataset to perform text classification with high accuracy.
Model Configuration
- Model Name: AlGe AIS
- Model Type: DeBERTa-v3
- Transformers Version: 4.21.3
Model Parameters
- Hidden Size: 1024
- Intermediate Size: 4096
- Number of Hidden Layers: 24
- Number of Attention Heads: 16
- Attention Dropout Probability: 0.1
- Hidden Dropout Probability: 0.1
- Hidden Activation Function: GELU
- Pooler Hidden Size: 1024
- Pooler Dropout Probability: 0
- Layer Normalization Epsilon: 1e-07
- Position Biased Input: False
- Maximum Position Embeddings: 512
- Maximum Relative Positions: -1
- Position Attention Types: p2c, c2p
- Relative Attention: True
- Share Attention Key: True
- Normalization of Relative Embeddings: Layer Normalization
- Vocabulary Size: 128100
- Padding Token ID: 0
- Type Vocabulary Size: 0
- Torch Data Type: float32
- Transformers Version: 4.21.3
Training Details
The model was trained on a specific dataset with the following settings:
31 Epochs
Evaluation Results
not final
| Metric | Score |
|---|---|
| MSE | 0.0111 |
| RMSE | 0.1055 |
| MAE | 0.0776 |
| R2 | 0.6485 |
| Cronbach's Alpha | 0.8937 |
Acknowledgments
This model was pretraine by the authors of DeBERTa-v3 and adapted for text classification tasks. We thank the authors for their contributions to the field of NLP and the Hugging Face team for providing the base DeBERTa-v3 model.
Disclaimer
The model card provides information about the specific configuration and training of the model. However, please note that the performance of the model may vary depending on the specific use case and input data. It is advisable to evaluate the model's performance in your specific context before deploying it in production.
- Downloads last month
- 0