<!-- 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. -->
add_BERT_24_stsb
This model is a fine-tuned version of gokuls/add_bert_12_layer_model_complete_training_new on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 1.4492
- Pearson: 0.6090
- Spearmanr: 0.6106
- Combined Score: 0.6098
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 |
---|---|---|---|---|---|---|
2.1466 | 1.0 | 45 | 2.2698 | 0.3930 | 0.3260 | 0.3595 |
1.4589 | 2.0 | 90 | 1.7726 | 0.5150 | 0.5141 | 0.5145 |
1.0006 | 3.0 | 135 | 1.4492 | 0.6090 | 0.6106 | 0.6098 |
0.6766 | 4.0 | 180 | 1.8200 | 0.5635 | 0.5672 | 0.5654 |
0.4849 | 5.0 | 225 | 2.1591 | 0.5213 | 0.5212 | 0.5213 |
0.3823 | 6.0 | 270 | 1.8541 | 0.5717 | 0.5716 | 0.5717 |
0.3158 | 7.0 | 315 | 1.8647 | 0.5777 | 0.5741 | 0.5759 |
0.2816 | 8.0 | 360 | 1.4985 | 0.6097 | 0.6083 | 0.6090 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.14.0a0+410ce96
- Datasets 2.13.0
- Tokenizers 0.13.3