metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_km_20_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
- f1
model-index:
- name: tiny_bert_km_20_v1_qqp
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE QQP
type: glue
args: qqp
metrics:
- name: Accuracy
type: accuracy
value: 0.8135790254761316
- name: F1
type: f1
value: 0.7354788895518197
tiny_bert_km_20_v1_qqp
This model is a fine-tuned version of Hartunka/tiny_bert_km_20_v1 on the GLUE QQP dataset. It achieves the following results on the evaluation set:
- Loss: 0.4122
- Accuracy: 0.8136
- F1: 0.7355
- Combined Score: 0.7745
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: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- seed: 10
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Combined Score |
|---|---|---|---|---|---|---|
| 0.4973 | 1.0 | 1422 | 0.4508 | 0.7838 | 0.6865 | 0.7351 |
| 0.409 | 2.0 | 2844 | 0.4178 | 0.8020 | 0.7201 | 0.7611 |
| 0.35 | 3.0 | 4266 | 0.4122 | 0.8136 | 0.7355 | 0.7745 |
| 0.3027 | 4.0 | 5688 | 0.4234 | 0.8185 | 0.7412 | 0.7798 |
| 0.2638 | 5.0 | 7110 | 0.4397 | 0.8230 | 0.7568 | 0.7899 |
| 0.2301 | 6.0 | 8532 | 0.4462 | 0.8205 | 0.7612 | 0.7909 |
| 0.2039 | 7.0 | 9954 | 0.4793 | 0.8194 | 0.7634 | 0.7914 |
| 0.1794 | 8.0 | 11376 | 0.5060 | 0.8209 | 0.7645 | 0.7927 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1