metadata
library_name: transformers
language:
- en
base_model: Hartunka/bert_base_km_100_v2
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: bert_base_km_100_v2_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE STSB
type: glue
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.22813141234967038
bert_base_km_100_v2_stsb
This model is a fine-tuned version of Hartunka/bert_base_km_100_v2 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.2511
- Pearson: 0.2543
- Spearmanr: 0.2281
- Combined Score: 0.2412
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 |
|---|---|---|---|---|---|---|
| 2.8847 | 1.0 | 23 | 2.3409 | 0.1427 | 0.1356 | 0.1391 |
| 1.9403 | 2.0 | 46 | 2.2677 | 0.2335 | 0.2099 | 0.2217 |
| 1.717 | 3.0 | 69 | 2.2511 | 0.2543 | 0.2281 | 0.2412 |
| 1.4331 | 4.0 | 92 | 2.4732 | 0.2501 | 0.2236 | 0.2368 |
| 1.1074 | 5.0 | 115 | 2.5302 | 0.2919 | 0.2773 | 0.2846 |
| 0.8742 | 6.0 | 138 | 2.2740 | 0.3420 | 0.3463 | 0.3442 |
| 0.6435 | 7.0 | 161 | 2.4854 | 0.3284 | 0.3262 | 0.3273 |
| 0.5047 | 8.0 | 184 | 2.6081 | 0.3154 | 0.3114 | 0.3134 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1