metadata
language:
- en
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: hBERTv2_new_pretrain_48_KD_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: GLUE STSB
type: glue
config: stsb
split: validation
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.23740661914182784
hBERTv2_new_pretrain_48_KD_stsb
This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_48_KD on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.2142
- Pearson: 0.2484
- Spearmanr: 0.2374
- Combined Score: 0.2429
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: 4e-05
- train_batch_size: 128
- eval_batch_size: 128
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 2.3019 | 1.0 | 45 | 2.6041 | 0.1631 | 0.1541 | 0.1586 |
| 2.1689 | 2.0 | 90 | 2.4035 | 0.1451 | 0.1550 | 0.1500 |
| 1.8521 | 3.0 | 135 | 2.2142 | 0.2484 | 0.2374 | 0.2429 |
| 1.6095 | 4.0 | 180 | 2.6155 | 0.2618 | 0.2566 | 0.2592 |
| 1.3325 | 5.0 | 225 | 2.9218 | 0.3160 | 0.3098 | 0.3129 |
| 1.0639 | 6.0 | 270 | 2.2315 | 0.3486 | 0.3476 | 0.3481 |
| 0.8434 | 7.0 | 315 | 2.5364 | 0.3459 | 0.3343 | 0.3401 |
| 0.7193 | 8.0 | 360 | 2.3979 | 0.3661 | 0.3633 | 0.3647 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3