Downloads · 30 days
10
18% of all-time downloads
hBERTv2_new_pretrain_48_ver2_stsb
gokuls/hBERTv2_new_pretrain_48_ver2_stsb
hBERTv2_new_pretrain_48_ver2_stsb is a text classification model from gokuls. Use it when you need a label for a piece of text. It is set up for transformers.
should probably proofread and complete it, then remove this comment. --
All-time downloads
57
Public
Repo size
5.2 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin477 MB · 100%
At a glance
- Task
- Text Classification
- Library
- transformers
- Model type
- hybridbert
- Access
- Public
- Created
- Oct 18, 2023
- Updated
- Oct 18, 2023
- SHA
- 89f6c2b6
Try a prompt
Base models
Datasets
Model card
From the Hugging Face model README
<!-- 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_ver2_stsb
This model is a fine-tuned version of 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: 2.0659
- Pearson: 0.3927
- Spearmanr: 0.3829
- Combined Score: 0.3878
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: 64
- eval_batch_size: 64
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 15
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 2.3158 | 1.0 | 90 | 2.7785 | 0.1355 | 0.1181 | 0.1268 |
| 2.0193 | 2.0 | 180 | 2.6566 | 0.1954 | 0.2050 | 0.2002 |
| 1.7471 | 3.0 | 270 | 2.5202 | 0.2578 | 0.2647 | 0.2612 |
| 1.5032 | 4.0 | 360 | 3.0601 | 0.2605 | 0.2732 | 0.2668 |
| 1.1825 | 5.0 | 450 | 2.3378 | 0.3150 | 0.3180 | 0.3165 |
| 0.8788 | 6.0 | 540 | 2.3657 | 0.3437 | 0.3421 | 0.3429 |
| 0.6987 | 7.0 | 630 | 2.0659 | 0.3927 | 0.3829 | 0.3878 |
| 0.5879 | 8.0 | 720 | 2.6712 | 0.3631 | 0.3636 | 0.3634 |
| 0.4865 | 9.0 | 810 | 2.3066 | 0.3665 | 0.3625 | 0.3645 |
| 0.4233 | 10.0 | 900 | 2.2781 | 0.3753 | 0.3695 | 0.3724 |
| 0.3628 | 11.0 | 990 | 2.4672 | 0.3848 | 0.3758 | 0.3803 |
| 0.3113 | 12.0 | 1080 | 2.4339 | 0.3873 | 0.3809 | 0.3841 |
Framework versions
- Transformers 4.34.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.14.5
- Tokenizers 0.14.1
- Task
- Text Classification
- Library
- transformers
- Type
- hybridbert
- Languages
- en
- Created
- Oct 18, 2023
- Updated
- Oct 18, 2023