<!-- 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_96
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: 0.6239
- Accuracy: 0.6838
- F1: 0.8122
- Combined Score: 0.7480
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 |
---|---|---|---|---|---|---|
0.6686 | 1.0 | 15 | 0.6467 | 0.6838 | 0.8122 | 0.7480 |
0.6433 | 2.0 | 30 | 0.6372 | 0.6838 | 0.8122 | 0.7480 |
0.6378 | 3.0 | 45 | 0.6319 | 0.6838 | 0.8122 | 0.7480 |
0.6344 | 4.0 | 60 | 0.6284 | 0.6838 | 0.8122 | 0.7480 |
0.6343 | 5.0 | 75 | 0.6266 | 0.6838 | 0.8122 | 0.7480 |
0.6299 | 6.0 | 90 | 0.6252 | 0.6838 | 0.8122 | 0.7480 |
0.6335 | 7.0 | 105 | 0.6247 | 0.6838 | 0.8122 | 0.7480 |
0.6308 | 8.0 | 120 | 0.6243 | 0.6838 | 0.8122 | 0.7480 |
0.6306 | 9.0 | 135 | 0.6243 | 0.6838 | 0.8122 | 0.7480 |
0.6302 | 10.0 | 150 | 0.6241 | 0.6838 | 0.8122 | 0.7480 |
0.6296 | 11.0 | 165 | 0.6241 | 0.6838 | 0.8122 | 0.7480 |
0.6305 | 12.0 | 180 | 0.6239 | 0.6838 | 0.8122 | 0.7480 |
0.634 | 13.0 | 195 | 0.6242 | 0.6838 | 0.8122 | 0.7480 |
0.63 | 14.0 | 210 | 0.6243 | 0.6838 | 0.8122 | 0.7480 |
0.6314 | 15.0 | 225 | 0.6242 | 0.6838 | 0.8122 | 0.7480 |
0.6286 | 16.0 | 240 | 0.6239 | 0.6838 | 0.8122 | 0.7480 |
0.6326 | 17.0 | 255 | 0.6242 | 0.6838 | 0.8122 | 0.7480 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2