final-lr2e-5-bs16-fp16-2
This model is a fine-tuned version of bert-base-uncased on an https://github.com/rewire-online/edos dataset. It achieves the following results on the evaluation set:
- Loss: 0.4823
- F1 Macro: 0.8301
- F1 Weighted: 0.8772
- F1: 0.7388
- Accuracy: 0.8792
- Confusion Matrix: [[2834 196] [ 287 683]]
- Confusion Matrix Norm: [[0.93531353 0.06468647] [0.29587629 0.70412371]]
- Classification Report: precision recall f1-score support 0 0.908042 0.935314 0.921476 3030.00000
1 0.777019 0.704124 0.738778 970.00000 accuracy 0.879250 0.879250 0.879250 0.87925 macro avg 0.842531 0.819719 0.830127 4000.00000 weighted avg 0.876269 0.879250 0.877172 4000.00000
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: 12345
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | F1 Macro | F1 Weighted | F1 | Accuracy | Confusion Matrix | Confusion Matrix Norm | Classification Report |
---|---|---|---|---|---|---|---|---|---|---|
0.3333 | 1.0 | 1000 | 0.3064 | 0.8165 | 0.8672 | 0.7181 | 0.8692 | [[2811 219] | ||
[ 304 666]] | [[0.92772277 0.07227723] | |||||||||
[0.31340206 0.68659794]] | precision recall f1-score support | |||||||||
0 0.902408 0.927723 0.914890 3030.00000 | ||||||||||
1 0.752542 0.686598 0.718059 970.00000 | ||||||||||
accuracy 0.869250 0.869250 0.869250 0.86925 | ||||||||||
macro avg 0.827475 0.807160 0.816475 4000.00000 | ||||||||||
weighted avg 0.866065 0.869250 0.867159 4000.00000 | ||||||||||
0.2271 | 2.0 | 2000 | 0.3905 | 0.8238 | 0.8708 | 0.7326 | 0.871 | [[2777 253] | ||
[ 263 707]] | [[0.91650165 0.08349835] | |||||||||
[0.27113402 0.72886598]] | precision recall f1-score support | |||||||||
0 0.913487 0.916502 0.914992 3030.000 | ||||||||||
1 0.736458 0.728866 0.732642 970.000 | ||||||||||
accuracy 0.871000 0.871000 0.871000 0.871 | ||||||||||
macro avg 0.824973 0.822684 0.823817 4000.000 | ||||||||||
weighted avg 0.870557 0.871000 0.870772 4000.000 | ||||||||||
0.1435 | 3.0 | 3000 | 0.4823 | 0.8301 | 0.8772 | 0.7388 | 0.8792 | [[2834 196] | ||
[ 287 683]] | [[0.93531353 0.06468647] | |||||||||
[0.29587629 0.70412371]] | precision recall f1-score support | |||||||||
0 0.908042 0.935314 0.921476 3030.00000 | ||||||||||
1 0.777019 0.704124 0.738778 970.00000 | ||||||||||
accuracy 0.879250 0.879250 0.879250 0.87925 | ||||||||||
macro avg 0.842531 0.819719 0.830127 4000.00000 | ||||||||||
weighted avg 0.876269 0.879250 0.877172 4000.00000 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu117
- Datasets 2.9.0
- Tokenizers 0.13.2
- Downloads last month
- 13
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.
Model tree for clincolnoz/bert-base-uncased-edos
Base model
google-bert/bert-base-uncased