metadata
library_name: transformers
base_model: Hartunka/tiny_bert_rand_20_v2
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: tiny_bert_rand_20_v2_mrpc
results: []
tiny_bert_rand_20_v2_mrpc
This model is a fine-tuned version of Hartunka/tiny_bert_rand_20_v2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.9225
- Accuracy: 0.6373
- F1: 0.7176
- Combined Score: 0.6774
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 |
|---|---|---|---|---|---|---|
| 0.6287 | 1.0 | 15 | 0.6033 | 0.7059 | 0.8187 | 0.7623 |
| 0.5917 | 2.0 | 30 | 0.5912 | 0.6936 | 0.8062 | 0.7499 |
| 0.5567 | 3.0 | 45 | 0.6027 | 0.6887 | 0.8013 | 0.7450 |
| 0.5147 | 4.0 | 60 | 0.6323 | 0.6765 | 0.7591 | 0.7178 |
| 0.4186 | 5.0 | 75 | 0.6771 | 0.6691 | 0.7550 | 0.7121 |
| 0.3291 | 6.0 | 90 | 0.7957 | 0.6716 | 0.7528 | 0.7122 |
| 0.242 | 7.0 | 105 | 0.9225 | 0.6373 | 0.7176 | 0.6774 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1