<!-- 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. -->
FLAN-T5_GLUE_finetuning_lr5e-5
This model is a fine-tuned version of google/flan-t5-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0797
- Accuracy: 0.8794
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5.0
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.141 | 0.17 | 2500 | 0.1083 | 0.8363 |
0.1096 | 0.34 | 5000 | 0.0986 | 0.8484 |
0.1029 | 0.51 | 7500 | 0.0919 | 0.8548 |
0.0994 | 0.68 | 10000 | 0.0895 | 0.8591 |
0.0956 | 0.85 | 12500 | 0.0871 | 0.8622 |
0.0938 | 1.02 | 15000 | 0.0862 | 0.8628 |
0.0849 | 1.18 | 17500 | 0.0845 | 0.8674 |
0.0845 | 1.35 | 20000 | 0.0849 | 0.8675 |
0.0837 | 1.52 | 22500 | 0.0835 | 0.8669 |
0.0826 | 1.69 | 25000 | 0.0838 | 0.871 |
0.0827 | 1.86 | 27500 | 0.0821 | 0.8711 |
0.0812 | 2.03 | 30000 | 0.0826 | 0.8709 |
0.0744 | 2.2 | 32500 | 0.0809 | 0.8747 |
0.0746 | 2.37 | 35000 | 0.0830 | 0.8728 |
0.0734 | 2.54 | 37500 | 0.0813 | 0.8741 |
0.0747 | 2.71 | 40000 | 0.0798 | 0.8755 |
0.0733 | 2.88 | 42500 | 0.0799 | 0.8753 |
0.0721 | 3.05 | 45000 | 0.0816 | 0.8755 |
0.0678 | 3.22 | 47500 | 0.0810 | 0.8774 |
0.0673 | 3.38 | 50000 | 0.0820 | 0.8759 |
0.0674 | 3.55 | 52500 | 0.0796 | 0.8774 |
0.0681 | 3.72 | 55000 | 0.0786 | 0.8783 |
0.0675 | 3.89 | 57500 | 0.0792 | 0.8785 |
0.066 | 4.06 | 60000 | 0.0794 | 0.879 |
0.0631 | 4.23 | 62500 | 0.0808 | 0.879 |
0.0638 | 4.4 | 65000 | 0.0804 | 0.8783 |
0.064 | 4.57 | 67500 | 0.0801 | 0.879 |
0.0637 | 4.74 | 70000 | 0.0801 | 0.8791 |
0.064 | 4.91 | 72500 | 0.0797 | 0.8794 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.0+cu117
- Datasets 2.10.1
- Tokenizers 0.12.1