metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- glue
metrics:
- spearmanr
model-index:
- name: mobilebert_sa_GLUE_Experiment_stsb
results:
- task:
name: Text Classification
type: text-classification
dataset:
name: glue
type: glue
config: stsb
split: validation
args: stsb
metrics:
- name: Spearmanr
type: spearmanr
value: 0.20056583736129271
mobilebert_sa_GLUE_Experiment_stsb
This model is a fine-tuned version of google/mobilebert-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 2.3885
- Pearson: 0.1998
- Spearmanr: 0.2006
- Combined Score: 0.2002
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: 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
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 4.0455 | 1.0 | 45 | 2.3024 | 0.0671 | 0.0815 | 0.0743 |
| 2.1712 | 2.0 | 90 | 2.6644 | 0.0612 | 0.0724 | 0.0668 |
| 2.0637 | 3.0 | 135 | 2.3625 | 0.0582 | 0.0669 | 0.0625 |
| 1.996 | 4.0 | 180 | 2.8671 | 0.0713 | 0.0728 | 0.0720 |
| 1.908 | 5.0 | 225 | 2.6622 | 0.0954 | 0.0898 | 0.0926 |
| 1.7068 | 6.0 | 270 | 2.3885 | 0.1998 | 0.2006 | 0.2002 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2