metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_km_10_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: tiny_bert_km_10_v1_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE STSB
type: glue
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.13759207934874104
tiny_bert_km_10_v1_stsb
This model is a fine-tuned version of Hartunka/tiny_bert_km_10_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.2270
- Pearson: 0.1335
- Spearmanr: 0.1376
- Combined Score: 0.1356
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 | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 3.5394 | 1.0 | 23 | 2.2270 | 0.1335 | 0.1376 | 0.1356 |
| 2.1228 | 2.0 | 46 | 2.2859 | 0.1504 | 0.1751 | 0.1628 |
| 1.9666 | 3.0 | 69 | 2.3973 | 0.1837 | 0.1858 | 0.1847 |
| 1.849 | 4.0 | 92 | 2.5343 | 0.1894 | 0.1721 | 0.1808 |
| 1.6195 | 5.0 | 115 | 2.6200 | 0.2320 | 0.2260 | 0.2290 |
| 1.4318 | 6.0 | 138 | 2.5622 | 0.2510 | 0.2461 | 0.2486 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1