<!-- 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. -->
distilbert_sa_GLUE_Experiment_mnli
This model is a fine-tuned version of distilbert-base-uncased on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.7995
- Accuracy: 0.6565
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: 256
- eval_batch_size: 256
- 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 | Accuracy |
---|---|---|---|---|
0.9882 | 1.0 | 1534 | 0.9194 | 0.5707 |
0.8859 | 2.0 | 3068 | 0.8623 | 0.6074 |
0.8254 | 3.0 | 4602 | 0.8507 | 0.6187 |
0.7672 | 4.0 | 6136 | 0.8192 | 0.6343 |
0.7114 | 5.0 | 7670 | 0.8120 | 0.6508 |
0.6566 | 6.0 | 9204 | 0.8250 | 0.6511 |
0.6012 | 7.0 | 10738 | 0.8666 | 0.6463 |
0.543 | 8.0 | 12272 | 0.8760 | 0.6572 |
0.4849 | 9.0 | 13806 | 0.9465 | 0.6579 |
0.429 | 10.0 | 15340 | 0.9820 | 0.6493 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2