metadata
library_name: transformers
language:
- en
base_model: Hartunka/tiny_bert_rand_5_v1
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: tiny_bert_rand_5_v1_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE STSB
type: glue
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.28550740239161837
tiny_bert_rand_5_v1_stsb
This model is a fine-tuned version of Hartunka/tiny_bert_rand_5_v1 on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.3876
- Pearson: 0.2931
- Spearmanr: 0.2855
- Combined Score: 0.2893
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 |
|---|---|---|---|---|---|---|
| 3.4897 | 1.0 | 23 | 2.4595 | 0.0690 | 0.0582 | 0.0636 |
| 2.0096 | 2.0 | 46 | 2.6146 | 0.1248 | 0.1084 | 0.1166 |
| 1.8479 | 3.0 | 69 | 2.3951 | 0.1948 | 0.1792 | 0.1870 |
| 1.6023 | 4.0 | 92 | 2.3876 | 0.2931 | 0.2855 | 0.2893 |
| 1.3085 | 5.0 | 115 | 2.4778 | 0.2811 | 0.2712 | 0.2761 |
| 1.0396 | 6.0 | 138 | 2.4495 | 0.3129 | 0.3111 | 0.3120 |
| 0.8308 | 7.0 | 161 | 2.4751 | 0.3386 | 0.3393 | 0.3390 |
| 0.6749 | 8.0 | 184 | 2.5454 | 0.3256 | 0.3217 | 0.3237 |
| 0.574 | 9.0 | 207 | 2.6687 | 0.3171 | 0.3124 | 0.3148 |
Framework versions
- Transformers 4.50.2
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.21.1