metadata
library_name: transformers
license: apache-2.0
base_model: bert-base-uncased
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: doc-bert
results: []
bert-news-classification
This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9503
- Accuracy: 0.7559
- Macro F1: 0.6911
- Macro Precision: 0.6843
- Macro Recall: 0.6995
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: 32
- eval_batch_size: 32
- seed: 42
- 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
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 4
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 | Macro Precision | Macro Recall |
|---|---|---|---|---|---|---|---|
| 1.0377 | 1.0 | 5238 | 0.9572 | 0.7330 | 0.6684 | 0.6598 | 0.6846 |
| 0.7817 | 2.0 | 10476 | 0.8938 | 0.7496 | 0.6851 | 0.6783 | 0.6975 |
| 0.5684 | 3.0 | 15714 | 0.9109 | 0.7544 | 0.6920 | 0.6830 | 0.7053 |
| 0.4249 | 4.0 | 20952 | 0.9503 | 0.7559 | 0.6911 | 0.6843 | 0.6995 |
Framework versions
- Transformers 4.53.3
- Pytorch 2.6.0+cu124
- Datasets 4.4.1
- Tokenizers 0.21.2