metadata
license: mit
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- precision
- recall
base_model: microsoft/deberta-v3-base
model-index:
- name: deberta-v3-base-isarcasm
results:
- task:
type: text-classification
dataset:
name: iSarcasm
type: isarcasm
split: test
metrics:
- type: f1
value: 0.47887323943661975
name: f1
- type: accuracy
value: 0.8331454340473506
name: accuracy
- type: recall
value: 0.43312101910828027
name: recall
- type: precision
value: 0.5354330708661418
name: precision
deberta-v3-base-isarcasm
This model is a fine-tuned version of microsoft/deberta-v3-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3693
- Accuracy: 0.8331
- F1: 0.4789
- Precision: 0.5354
- Recall: 0.4331
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 215 | 0.7833 | 0.8 | 0.0 | 0.0 | 0.0 |
| No log | 2.0 | 430 | 1.1913 | 0.8 | 0.0 | 0.0 | 0.0 |
| 0.577 | 3.0 | 645 | 1.5866 | 0.7714 | 0.2 | 0.25 | 0.1667 |
| 0.577 | 4.0 | 860 | 2.3199 | 0.8 | 0.2222 | 0.3333 | 0.1667 |
| 0.2047 | 5.0 | 1075 | 2.4911 | 0.8 | 0.2222 | 0.3333 | 0.1667 |
Framework versions
- Transformers 4.32.0
- Pytorch 2.1.1+cu121
- Datasets 2.14.5
- Tokenizers 0.13.3