metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_km_10_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- accuracy
- f1
model-index:
- name: tiny_bert_km_10_v1_mrpc
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE MRPC
type: glue
args: mrpc
metrics:
- name: Accuracy
type: accuracy
value: 0.7009803921568627
- name: F1
type: f1
value: 0.8032258064516129
tiny_bert_km_10_v1_mrpc
This model is a fine-tuned version of Hartunka/tiny_bert_km_10_v1 on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.5893
- Accuracy: 0.7010
- F1: 0.8032
- Combined Score: 0.7521
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 |
|---|---|---|---|---|---|---|
| 0.6324 | 1.0 | 15 | 0.6063 | 0.7108 | 0.8223 | 0.7665 |
| 0.596 | 2.0 | 30 | 0.5934 | 0.7034 | 0.8191 | 0.7613 |
| 0.5663 | 3.0 | 45 | 0.5923 | 0.7083 | 0.8200 | 0.7642 |
| 0.5449 | 4.0 | 60 | 0.5893 | 0.7010 | 0.8032 | 0.7521 |
| 0.4957 | 5.0 | 75 | 0.6304 | 0.6569 | 0.75 | 0.7034 |
| 0.4356 | 6.0 | 90 | 0.6516 | 0.6936 | 0.7871 | 0.7403 |
| 0.3509 | 7.0 | 105 | 0.7439 | 0.6887 | 0.7908 | 0.7397 |
| 0.2548 | 8.0 | 120 | 0.8600 | 0.6618 | 0.7553 | 0.7085 |
| 0.1693 | 9.0 | 135 | 1.0956 | 0.6225 | 0.7148 | 0.6687 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1