---
language:
- en
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: hBERTv2_new_pretrain_48_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: 0.4028161409951644
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# hBERTv2_new_pretrain_48_stsb

This model is a fine-tuned version of [gokuls/bert_12_layer_model_v2_complete_training_new_48](https://huggingface.co/gokuls/bert_12_layer_model_v2_complete_training_new_48) on the GLUE STSB dataset.
It achieves the following results on the evaluation set:
- Loss: 2.0734
- Pearson: 0.4184
- Spearmanr: 0.4028
- Combined Score: 0.4106

## 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 |
|:-------------:|:-----:|:----:|:---------------:|:-------:|:---------:|:--------------:|
| 2.2864        | 1.0   | 45   | 3.0157          | 0.1270  | 0.1171    | 0.1220         |
| 1.9895        | 2.0   | 90   | 2.7270          | 0.1553  | 0.1550    | 0.1552         |
| 1.7101        | 3.0   | 135  | 2.8223          | 0.2806  | 0.2657    | 0.2732         |
| 1.2973        | 4.0   | 180  | 2.5938          | 0.3375  | 0.3280    | 0.3328         |
| 1.0658        | 5.0   | 225  | 2.3835          | 0.3771  | 0.3629    | 0.3700         |
| 0.8454        | 6.0   | 270  | 2.5028          | 0.3637  | 0.3479    | 0.3558         |
| 0.6773        | 7.0   | 315  | 2.3937          | 0.3594  | 0.3538    | 0.3566         |
| 0.5678        | 8.0   | 360  | 2.6813          | 0.3803  | 0.3802    | 0.3803         |
| 0.4746        | 9.0   | 405  | 2.5546          | 0.3874  | 0.3695    | 0.3784         |
| 0.4113        | 10.0  | 450  | 2.2077          | 0.4112  | 0.4038    | 0.4075         |
| 0.3585        | 11.0  | 495  | 2.2846          | 0.4096  | 0.3972    | 0.4034         |
| 0.3288        | 12.0  | 540  | 2.4155          | 0.4012  | 0.3848    | 0.3930         |
| 0.2745        | 13.0  | 585  | 2.3635          | 0.4004  | 0.3924    | 0.3964         |
| 0.2579        | 14.0  | 630  | 2.0734          | 0.4184  | 0.4028    | 0.4106         |
| 0.2309        | 15.0  | 675  | 2.3462          | 0.4171  | 0.4026    | 0.4099         |
| 0.2037        | 16.0  | 720  | 2.2598          | 0.4225  | 0.4090    | 0.4157         |
| 0.1806        | 17.0  | 765  | 2.2458          | 0.4116  | 0.3916    | 0.4016         |
| 0.1785        | 18.0  | 810  | 2.3296          | 0.4088  | 0.3903    | 0.3996         |
| 0.1582        | 19.0  | 855  | 2.3369          | 0.4033  | 0.3868    | 0.3951         |


### Framework versions

- Transformers 4.29.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.12.0
- Tokenizers 0.13.3