metadata
library_name: transformers
language:
- en
base_model: Hartunka/bert_base_km_10_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: bert_base_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: 0.20741798107513082
bert_base_km_10_v1_stsb
This model is a fine-tuned version of Hartunka/bert_base_km_10_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.1875
- Pearson: 0.2207
- Spearmanr: 0.2074
- Combined Score: 0.2141
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 |
|---|---|---|---|---|---|---|
| 2.7027 | 1.0 | 23 | 2.1875 | 0.2207 | 0.2074 | 0.2141 |
| 1.9228 | 2.0 | 46 | 2.1904 | 0.2334 | 0.2271 | 0.2303 |
| 1.6984 | 3.0 | 69 | 2.2392 | 0.2805 | 0.2828 | 0.2816 |
| 1.3717 | 4.0 | 92 | 2.3772 | 0.2851 | 0.2845 | 0.2848 |
| 1.044 | 5.0 | 115 | 2.4946 | 0.3093 | 0.3120 | 0.3107 |
| 0.7602 | 6.0 | 138 | 2.5728 | 0.2738 | 0.2716 | 0.2727 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1