| --- |
| license: mit |
| tags: |
| - generated_from_trainer |
| metrics: |
| - accuracy |
| model-index: |
| - name: SST2_XLNet_5E |
| results: [] |
| --- |
| |
| <!-- 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. --> |
|
|
| # SST2_XLNet_5E |
|
|
| This model is a fine-tuned version of [xlnet-base-cased](https://huggingface.co/xlnet-base-cased) on an unknown dataset. |
| It achieves the following results on the evaluation set: |
| - Loss: 0.5502 |
| - Accuracy: 0.9133 |
|
|
| ## 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: 8 |
| - seed: 42 |
| - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
| - lr_scheduler_type: linear |
| - num_epochs: 5 |
|
|
| ### Training results |
|
|
| | Training Loss | Epoch | Step | Validation Loss | Accuracy | |
| |:-------------:|:-----:|:----:|:---------------:|:--------:| |
| | 0.6038 | 0.12 | 50 | 0.2830 | 0.8933 | |
| | 0.3903 | 0.23 | 100 | 0.3346 | 0.9 | |
| | 0.3476 | 0.35 | 150 | 0.4187 | 0.8533 | |
| | 0.3528 | 0.46 | 200 | 0.3177 | 0.9 | |
| | 0.3372 | 0.58 | 250 | 0.4171 | 0.8333 | |
| | 0.3106 | 0.69 | 300 | 0.2825 | 0.9 | |
| | 0.295 | 0.81 | 350 | 0.3152 | 0.9 | |
| | 0.2828 | 0.92 | 400 | 0.4360 | 0.88 | |
| | 0.2359 | 1.04 | 450 | 0.3971 | 0.9 | |
| | 0.2224 | 1.15 | 500 | 0.3380 | 0.88 | |
| | 0.2136 | 1.27 | 550 | 0.3889 | 0.8933 | |
| | 0.264 | 1.39 | 600 | 0.4182 | 0.8667 | |
| | 0.1864 | 1.5 | 650 | 0.4887 | 0.88 | |
| | 0.1817 | 1.62 | 700 | 0.3626 | 0.9133 | |
| | 0.2021 | 1.73 | 750 | 0.4481 | 0.8933 | |
| | 0.2154 | 1.85 | 800 | 0.3702 | 0.8933 | |
| | 0.2392 | 1.96 | 850 | 0.5025 | 0.8933 | |
| | 0.1496 | 2.08 | 900 | 0.4606 | 0.9133 | |
| | 0.1537 | 2.19 | 950 | 0.5008 | 0.8933 | |
| | 0.1015 | 2.31 | 1000 | 0.5612 | 0.9067 | |
| | 0.0915 | 2.42 | 1050 | 0.5249 | 0.8933 | |
| | 0.1239 | 2.54 | 1100 | 0.4234 | 0.9133 | |
| | 0.1135 | 2.66 | 1150 | 0.4910 | 0.9067 | |
| | 0.1738 | 2.77 | 1200 | 0.3844 | 0.92 | |
| | 0.1428 | 2.89 | 1250 | 0.4282 | 0.92 | |
| | 0.1282 | 3.0 | 1300 | 0.4320 | 0.9 | |
| | 0.059 | 3.12 | 1350 | 0.4957 | 0.9133 | |
| | 0.0517 | 3.23 | 1400 | 0.4927 | 0.92 | |
| | 0.0853 | 3.35 | 1450 | 0.4187 | 0.92 | |
| | 0.0808 | 3.46 | 1500 | 0.4304 | 0.92 | |
| | 0.09 | 3.58 | 1550 | 0.3447 | 0.9267 | |
| | 0.044 | 3.7 | 1600 | 0.4994 | 0.9067 | |
| | 0.0443 | 3.81 | 1650 | 0.4516 | 0.9133 | |
| | 0.0974 | 3.93 | 1700 | 0.4172 | 0.92 | |
| | 0.0768 | 4.04 | 1750 | 0.4777 | 0.9133 | |
| | 0.0418 | 4.16 | 1800 | 0.4924 | 0.9267 | |
| | 0.0237 | 4.27 | 1850 | 0.5254 | 0.92 | |
| | 0.0426 | 4.39 | 1900 | 0.5532 | 0.9133 | |
| | 0.0336 | 4.5 | 1950 | 0.5838 | 0.9067 | |
| | 0.0188 | 4.62 | 2000 | 0.5775 | 0.9067 | |
| | 0.0318 | 4.73 | 2050 | 0.5781 | 0.9067 | |
| | 0.0348 | 4.85 | 2100 | 0.5526 | 0.9133 | |
| | 0.0524 | 4.97 | 2150 | 0.5502 | 0.9133 | |
|
|
|
|
| ### Framework versions |
|
|
| - Transformers 4.23.1 |
| - Pytorch 1.13.0 |
| - Datasets 2.6.1 |
| - Tokenizers 0.13.1 |
|
|
