metadata
license: mit
base_model: microsoft/deberta-v3-large
tags:
- generated_from_trainer
datasets:
- boolq
metrics:
- accuracy
model-index:
- name: deberta-v3-large_boolq
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: boolq
type: boolq
config: default
split: validation
args: default
metrics:
- name: Accuracy
type: accuracy
value: 0.882262996941896
deberta-v3-large_boolq
This model is a fine-tuned version of microsoft/deberta-v3-large on the boolq dataset. It achieves the following results on the evaluation set:
- Loss: 0.5796
- Accuracy: 0.8823
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 2
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 0.85 | 250 | 0.3265 | 0.8740 |
| 0.3786 | 1.69 | 500 | 0.3212 | 0.8844 |
| 0.3786 | 2.54 | 750 | 0.4205 | 0.8838 |
| 0.1324 | 3.39 | 1000 | 0.5393 | 0.8832 |
| 0.1324 | 4.24 | 1250 | 0.5796 | 0.8823 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.0.1+cu117
- Datasets 2.14.4
- Tokenizers 0.13.3