README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: bert_sm_cv_defined_summarized_4
    results: []

bert_sm_cv_defined_summarized_4

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: 1.8001
• Accuracy: 0.801
• Precision: 0.4677
• Recall: 0.1487
• F1: 0.2257
• D-index: 1.4847

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: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 8000
• num_epochs: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	D-index
No log	1.0	250	0.4931	0.805	0.5	0.0308	0.0580	1.4481
0.5724	2.0	500	0.4850	0.806	0.5263	0.0513	0.0935	1.4569
0.5724	3.0	750	0.4842	0.811	0.6	0.0923	0.16	1.4785
0.4468	4.0	1000	0.4954	0.81	0.5806	0.0923	0.1593	1.4771
0.4468	5.0	1250	0.5307	0.81	0.5862	0.0872	0.1518	1.4753
0.381	6.0	1500	0.5312	0.809	0.5455	0.1231	0.2008	1.4866
0.381	7.0	1750	0.5354	0.807	0.5161	0.1641	0.2490	1.4983
0.283	8.0	2000	0.7003	0.811	0.6364	0.0718	0.1290	1.4712
0.283	9.0	2250	0.7079	0.798	0.4568	0.1897	0.2681	1.4949
0.1621	10.0	2500	0.9032	0.8	0.4603	0.1487	0.2248	1.4833
0.1621	11.0	2750	1.0875	0.797	0.4474	0.1744	0.2509	1.4881
0.0678	12.0	3000	1.2256	0.769	0.3861	0.3128	0.3456	1.4975
0.0678	13.0	3250	1.6378	0.793	0.4	0.1231	0.1882	1.4645
0.039	14.0	3500	1.7475	0.767	0.2841	0.1282	0.1767	1.4301
0.039	15.0	3750	1.8575	0.804	0.4848	0.0821	0.1404	1.4652
0.0295	16.0	4000	1.8151	0.775	0.3370	0.1590	0.2160	1.4522
0.0295	17.0	4250	1.8788	0.795	0.4219	0.1385	0.2085	1.4728
0.0416	18.0	4500	1.8193	0.765	0.3462	0.2308	0.2769	1.4636
0.0416	19.0	4750	1.6942	0.788	0.3896	0.1538	0.2206	1.4685
0.0322	20.0	5000	1.8001	0.801	0.4677	0.1487	0.2257	1.4847

Framework versions

• Transformers 4.28.0
• Pytorch 2.0.1+cu118
• Datasets 2.12.0
• Tokenizers 0.13.3