metadata
license: mit
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: xlnet-base-cased_fold_1_binary_v1
results: []
xlnet-base-cased_fold_1_binary_v1
This model is a fine-tuned version of xlnet-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.7812
- F1: 0.8161
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: 2e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 25
Training results
| Training Loss | Epoch | Step | Validation Loss | F1 |
|---|---|---|---|---|
| No log | 1.0 | 288 | 0.3938 | 0.8019 |
| 0.4444 | 2.0 | 576 | 0.3945 | 0.8086 |
| 0.4444 | 3.0 | 864 | 0.4738 | 0.8245 |
| 0.2504 | 4.0 | 1152 | 0.6641 | 0.8123 |
| 0.2504 | 5.0 | 1440 | 0.8714 | 0.7863 |
| 0.159 | 6.0 | 1728 | 0.9177 | 0.8179 |
| 0.0832 | 7.0 | 2016 | 1.1719 | 0.8129 |
| 0.0832 | 8.0 | 2304 | 1.2858 | 0.8146 |
| 0.046 | 9.0 | 2592 | 1.2557 | 0.8181 |
| 0.046 | 10.0 | 2880 | 1.3332 | 0.8033 |
| 0.0313 | 11.0 | 3168 | 1.2840 | 0.8112 |
| 0.0313 | 12.0 | 3456 | 1.4164 | 0.8175 |
| 0.0246 | 13.0 | 3744 | 1.3709 | 0.8143 |
| 0.0173 | 14.0 | 4032 | 1.4319 | 0.8179 |
| 0.0173 | 15.0 | 4320 | 1.5706 | 0.8195 |
| 0.0138 | 16.0 | 4608 | 1.6072 | 0.8230 |
| 0.0138 | 17.0 | 4896 | 1.7454 | 0.8192 |
| 0.0016 | 18.0 | 5184 | 1.7281 | 0.8099 |
| 0.0016 | 19.0 | 5472 | 1.7692 | 0.8151 |
| 0.0088 | 20.0 | 5760 | 1.7376 | 0.8132 |
| 0.0081 | 21.0 | 6048 | 1.7715 | 0.8086 |
| 0.0081 | 22.0 | 6336 | 1.7400 | 0.8152 |
| 0.0053 | 23.0 | 6624 | 1.7845 | 0.8099 |
| 0.0053 | 24.0 | 6912 | 1.8096 | 0.8150 |
| 0.0062 | 25.0 | 7200 | 1.7812 | 0.8161 |
Framework versions
- Transformers 4.21.1
- Pytorch 1.12.0+cu113
- Datasets 2.4.0
- Tokenizers 0.12.1