<!-- 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_add_GLUE_Experiment_mrpc
This model is a fine-tuned version of distilbert-base-uncased on the GLUE MRPC dataset. It achieves the following results on the evaluation set:
- Loss: 0.6028
- Accuracy: 0.6961
- F1: 0.8171
- Combined Score: 0.7566
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 | F1 | Combined Score |
---|---|---|---|---|---|---|
0.6617 | 1.0 | 15 | 0.6507 | 0.6838 | 0.8122 | 0.7480 |
0.6412 | 2.0 | 30 | 0.6290 | 0.6838 | 0.8122 | 0.7480 |
0.6315 | 3.0 | 45 | 0.6252 | 0.6838 | 0.8122 | 0.7480 |
0.6319 | 4.0 | 60 | 0.6236 | 0.6838 | 0.8122 | 0.7480 |
0.6321 | 5.0 | 75 | 0.6225 | 0.6838 | 0.8122 | 0.7480 |
0.616 | 6.0 | 90 | 0.6028 | 0.6961 | 0.8171 | 0.7566 |
0.5469 | 7.0 | 105 | 0.6485 | 0.6446 | 0.7349 | 0.6898 |
0.4436 | 8.0 | 120 | 0.7536 | 0.6838 | 0.7909 | 0.7374 |
0.3794 | 9.0 | 135 | 0.7805 | 0.6961 | 0.7898 | 0.7430 |
0.3158 | 10.0 | 150 | 0.8811 | 0.6838 | 0.7825 | 0.7331 |
0.281 | 11.0 | 165 | 0.9246 | 0.6863 | 0.7881 | 0.7372 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2