Text Classification
Transformers
PyTorch
TensorBoard
deberta-v2
Generated from Trainer
Eval Results (legacy)
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("Emanuel/twitter-emotion-deberta-v3-base")
model = AutoModelForSequenceClassification.from_pretrained("Emanuel/twitter-emotion-deberta-v3-base", device_map="auto")Quick Links
twitter-emotion-deberta-v3-base
This model is a fine-tuned version of DeBERTa-v3. It achieves the following results on the evaluation set:
- Loss: 0.1474
- Accuracy: 0.937
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 80
- eval_batch_size: 80
- lr_scheduler_type: linear
- num_epochs: 6.0
Framework versions
- Transformers 4.12.5
- Pytorch 1.10.0+cu113
- Datasets 1.15.1
- Tokenizers 0.10.3
- Downloads last month
- 31
Dataset used to train Emanuel/twitter-emotion-deberta-v3-base
Spaces using Emanuel/twitter-emotion-deberta-v3-base 2
Evaluation results
- Accuracy on emotionself-reported0.937
- Accuracy on emotiontest set verified0.925
- Precision Macro on emotiontest set verified0.892
- Precision Micro on emotiontest set verified0.925
- Precision Weighted on emotiontest set verified0.929
- Recall Macro on emotiontest set verified0.876
- Recall Micro on emotiontest set verified0.925
- Recall Weighted on emotiontest set verified0.925
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="Emanuel/twitter-emotion-deberta-v3-base")