<!-- 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. -->
org_and_korquad
This model is a fine-tuned version of klue/roberta-large on the None dataset. It achieves the following results on the evaluation set:
- Exact Match: 69.1667
- F1: 77.4615
- Loss: 1.8961
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: 3e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10.0
Training results
Training Loss | Epoch | Step | Exact Match | F1 | Validation Loss |
---|---|---|---|---|---|
No log | 0.83 | 500 | 62.5 | 70.6430 | 0.9231 |
No log | 1.65 | 1000 | 66.25 | 75.8942 | 0.8221 |
No log | 2.48 | 1500 | 67.9167 | 76.0458 | 0.9581 |
No log | 3.31 | 2000 | 69.1667 | 77.5276 | 1.1324 |
No log | 4.14 | 2500 | 67.9167 | 76.8019 | 1.2806 |
No log | 4.97 | 3000 | 68.75 | 76.8566 | 1.3316 |
No log | 5.79 | 3500 | 66.25 | 74.1136 | 1.4468 |
No log | 6.62 | 4000 | 65.0 | 73.9254 | 1.5662 |
No log | 7.45 | 4500 | 67.9167 | 76.6298 | 1.5423 |
No log | 8.28 | 5000 | 67.5 | 75.6113 | 1.7773 |
No log | 9.11 | 5500 | 68.3333 | 76.5557 | 1.8478 |
No log | 9.93 | 6000 | 69.1667 | 77.4615 | 1.8961 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.13.0+cu117
- Datasets 2.7.1
- Tokenizers 0.13.2