metadata
library_name: peft
license: apache-2.0
base_model: google-bert/bert-base-cased
tags:
- generated_from_trainer
model-index:
- name: grandiose-horse-172
results: []
grandiose-horse-172
This model is a fine-tuned version of google-bert/bert-base-cased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6509
- Hamming Loss: 0.3414
- Zero One Loss: 1.0
- Jaccard Score: 0.8678
- Hamming Loss Optimised: 0.1121
- Hamming Loss Threshold: 0.7504
- Zero One Loss Optimised: 0.8812
- Zero One Loss Threshold: 0.6730
- Jaccard Score Optimised: 0.8449
- Jaccard Score Threshold: 0.6539
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: 1.510606094120106e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 2024
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Hamming Loss | Zero One Loss | Jaccard Score | Hamming Loss Optimised | Hamming Loss Threshold | Zero One Loss Optimised | Zero One Loss Threshold | Jaccard Score Optimised | Jaccard Score Threshold |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| No log | 1.0 | 100 | 0.7202 | 0.4325 | 1.0 | 0.8586 | 0.1123 | 0.7924 | 0.8712 | 0.7112 | 0.8203 | 0.5766 |
| No log | 2.0 | 200 | 0.6922 | 0.3761 | 1.0 | 0.8520 | 0.1123 | 0.7829 | 0.8812 | 0.6982 | 0.8546 | 0.5904 |
| No log | 3.0 | 300 | 0.6696 | 0.349 | 1.0 | 0.8606 | 0.1123 | 0.7641 | 0.885 | 0.6857 | 0.8436 | 0.6634 |
| No log | 4.0 | 400 | 0.6555 | 0.3432 | 1.0 | 0.8662 | 0.1121 | 0.7518 | 0.8825 | 0.6757 | 0.8455 | 0.6604 |
| 0.6931 | 5.0 | 500 | 0.6509 | 0.3414 | 1.0 | 0.8678 | 0.1121 | 0.7504 | 0.8812 | 0.6730 | 0.8449 | 0.6539 |
Framework versions
- PEFT 0.13.2
- Transformers 4.47.0
- Pytorch 2.5.1+cu124
- Datasets 3.1.0
- Tokenizers 0.21.0