README.md

metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - crows_pairs
metrics:
  - accuracy
model-index:
  - name: bert-base-uncased_crows_pairs_classifieronly
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: crows_pairs
          type: crows_pairs
          config: crows_pairs
          split: test
          args: crows_pairs
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.4735099337748344

bert-base-uncased_crows_pairs_classifieronly

This model is a fine-tuned version of bert-base-uncased on the crows_pairs dataset. It achieves the following results on the evaluation set:

• Loss: 0.6928
• Accuracy: 0.4735
• Tp: 0.3278
• Tn: 0.1457
• Fp: 0.3179
• Fn: 0.2086

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: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Tp	Tn	Fp	Fn
0.7033	1.05	20	0.6951	0.4636	0.0	0.4636	0.0	0.5364
0.7008	2.11	40	0.6951	0.4636	0.0	0.4636	0.0	0.5364
0.6998	3.16	60	0.6939	0.4570	0.0	0.4570	0.0066	0.5364
0.6958	4.21	80	0.6948	0.4636	0.0	0.4636	0.0	0.5364
0.7025	5.26	100	0.6981	0.4636	0.0	0.4636	0.0	0.5364
0.7083	6.32	120	0.6917	0.5397	0.5364	0.0033	0.4603	0.0
0.7016	7.37	140	0.6939	0.4570	0.0	0.4570	0.0066	0.5364
0.7061	8.42	160	0.6984	0.4636	0.0	0.4636	0.0	0.5364
0.698	9.47	180	0.6948	0.4636	0.0	0.4636	0.0	0.5364
0.7009	10.53	200	0.6931	0.4967	0.1623	0.3344	0.1291	0.3742
0.7047	11.58	220	0.6998	0.4636	0.0	0.4636	0.0	0.5364
0.6945	12.63	240	0.6935	0.4868	0.0364	0.4503	0.0132	0.5
0.708	13.68	260	0.6924	0.5364	0.5033	0.0331	0.4305	0.0331
0.7007	14.74	280	0.6935	0.4834	0.0331	0.4503	0.0132	0.5033
0.6999	15.79	300	0.6938	0.4636	0.0066	0.4570	0.0066	0.5298
0.6993	16.84	320	0.6939	0.4570	0.0	0.4570	0.0066	0.5364
0.7002	17.89	340	0.6953	0.4636	0.0	0.4636	0.0	0.5364
0.7025	18.95	360	0.6956	0.4636	0.0	0.4636	0.0	0.5364
0.7008	20.0	380	0.6905	0.5397	0.5364	0.0033	0.4603	0.0
0.7065	21.05	400	0.6970	0.4636	0.0	0.4636	0.0	0.5364
0.6996	22.11	420	0.6954	0.4636	0.0	0.4636	0.0	0.5364
0.7	23.16	440	0.6962	0.4636	0.0	0.4636	0.0	0.5364
0.7028	24.21	460	0.6948	0.4636	0.0	0.4636	0.0	0.5364
0.6924	25.26	480	0.6930	0.4834	0.2020	0.2815	0.1821	0.3344
0.6973	26.32	500	0.6941	0.4636	0.0	0.4636	0.0	0.5364
0.6953	27.37	520	0.6938	0.4603	0.0033	0.4570	0.0066	0.5331
0.6971	28.42	540	0.6928	0.4735	0.3411	0.1325	0.3311	0.1954
0.7086	29.47	560	0.6924	0.5199	0.5066	0.0132	0.4503	0.0298
0.6959	30.53	580	0.6925	0.5066	0.4636	0.0430	0.4205	0.0728
0.7103	31.58	600	0.6919	0.5397	0.5364	0.0033	0.4603	0.0
0.7019	32.63	620	0.6916	0.5397	0.5364	0.0033	0.4603	0.0
0.6941	33.68	640	0.6935	0.4868	0.0364	0.4503	0.0132	0.5
0.6878	34.74	660	0.6959	0.4636	0.0	0.4636	0.0	0.5364
0.6995	35.79	680	0.6954	0.4636	0.0	0.4636	0.0	0.5364
0.6968	36.84	700	0.6916	0.5397	0.5364	0.0033	0.4603	0.0
0.6997	37.89	720	0.6921	0.5331	0.5265	0.0066	0.4570	0.0099
0.6975	38.95	740	0.6964	0.4636	0.0	0.4636	0.0	0.5364
0.7026	40.0	760	0.6956	0.4636	0.0	0.4636	0.0	0.5364
0.7057	41.05	780	0.6943	0.4636	0.0	0.4636	0.0	0.5364
0.7028	42.11	800	0.6953	0.4636	0.0	0.4636	0.0	0.5364
0.6987	43.16	820	0.6938	0.4603	0.0033	0.4570	0.0066	0.5331
0.6973	44.21	840	0.6933	0.4868	0.0497	0.4371	0.0265	0.4868
0.7119	45.26	860	0.6930	0.4801	0.1788	0.3013	0.1623	0.3576
0.7041	46.32	880	0.6928	0.4967	0.3179	0.1788	0.2848	0.2185
0.7114	47.37	900	0.6926	0.4967	0.4139	0.0828	0.3808	0.1225
0.702	48.42	920	0.6929	0.4735	0.2318	0.2417	0.2219	0.3046
0.6945	49.47	940	0.6928	0.4735	0.3278	0.1457	0.3179	0.2086

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1
• Datasets 2.10.1
• Tokenizers 0.13.2