metadata
license: mit
base_model: deepset/gbert-base
tags:
- generated_from_trainer
metrics:
- f1
model-index:
- name: gbert-base
results: []
gbert-base
This model is a fine-tuned version of deepset/gbert-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.1992
- F1: 0.4000
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: 4e-06
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | F1 |
---|---|---|---|---|
0.6664 | 1.0 | 189 | 0.6660 | 0.0 |
0.5942 | 2.0 | 378 | 0.5671 | 0.3750 |
0.5593 | 3.0 | 567 | 0.8488 | 0.3111 |
0.6282 | 4.0 | 756 | 0.6896 | 0.4444 |
0.558 | 5.0 | 945 | 0.8825 | 0.4167 |
0.5242 | 6.0 | 1134 | 1.0105 | 0.3774 |
0.4843 | 7.0 | 1323 | 1.1065 | 0.4082 |
0.3792 | 8.0 | 1512 | 1.1807 | 0.4 |
0.4005 | 9.0 | 1701 | 1.2279 | 0.3774 |
0.3909 | 10.0 | 1890 | 1.1992 | 0.4000 |
Framework versions
- Transformers 4.32.1
- Pytorch 2.1.2
- Datasets 2.12.0
- Tokenizers 0.13.3