<!-- 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. -->
tabert-1k-indic_glue
This model is a fine-tuned version of livinNector/tabert-1k on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.2633
- Precision: 0.8087
- Recall: 0.8327
- F1: 0.8205
- Accuracy: 0.9183
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: 32
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 2
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
---|---|---|---|---|---|---|---|
0.5381 | 0.31 | 200 | 0.3586 | 0.7418 | 0.7279 | 0.7348 | 0.8813 |
0.3592 | 0.62 | 400 | 0.3274 | 0.7271 | 0.7984 | 0.7611 | 0.8917 |
0.2984 | 0.94 | 600 | 0.2852 | 0.7870 | 0.7958 | 0.7914 | 0.9058 |
0.2299 | 1.25 | 800 | 0.2819 | 0.7918 | 0.8158 | 0.8036 | 0.9118 |
0.1841 | 1.56 | 1000 | 0.2651 | 0.8143 | 0.8168 | 0.8155 | 0.9158 |
0.1767 | 1.88 | 1200 | 0.2633 | 0.8087 | 0.8327 | 0.8205 | 0.9183 |
Framework versions
- Transformers 4.29.2
- Pytorch 2.0.0+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3