claim-judge

This model is a fine-tuned version of bert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

• Accuracy: 0.7623
• Loss: 1.5465

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: 160
• eval_batch_size: 160
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 36

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
0.4968	1.0	1041	0.7587	0.6138
0.4174	2.0	2082	0.7695	0.6211
0.4036	3.0	3123	0.7733	0.6181
0.3415	4.0	4164	0.7772	0.6450
0.2868	5.0	5205	0.7738	0.6896
0.2453	6.0	6246	0.7763	0.7119
0.2003	7.0	7287	0.7728	0.8254
0.1683	8.0	8328	0.7712	0.9288
0.1439	9.0	9369	0.7764	0.8993
0.1197	10.0	10410	0.7729	0.9819
0.102	11.0	11451	0.7709	1.0478
0.088	12.0	12492	0.7692	1.1574
0.087	13.0	13533	0.7709	1.0969
0.0779	14.0	14574	0.7661	1.2575
0.0695	15.0	15615	0.7658	1.3540
0.0664	16.0	16656	0.7719	1.2155
0.058	17.0	17697	0.7654	1.3065
0.0533	18.0	18738	0.7674	1.3535
0.0496	19.0	19779	0.7663	1.3327
0.0459	20.0	20820	0.7686	1.3893
0.0432	21.0	21861	0.7691	1.4211
0.0396	22.0	22902	0.7682	1.4810
0.0371	23.0	23943	0.7705	1.4926
0.0338	24.0	24984	0.7633	1.5058
0.037	25.0	26025	0.7604	1.4986
0.034	26.0	27066	0.7611	1.5314
0.0317	27.0	28107	0.7659	1.4636
0.0312	28.0	29148	0.7658	1.5006
0.0282	29.0	30189	0.7672	1.4250
0.0282	30.0	31230	0.7662	1.4904
0.0264	31.0	32271	0.7669	1.5415
0.0253	32.0	33312	0.7679	1.6110
0.0257	33.0	34353	0.7645	1.6097
0.0233	34.0	35394	0.7614	1.6646
0.0251	35.0	36435	0.7651	1.6080
0.0236	36.0	37476	0.7699	1.5824
0.025	37.0	38517	0.7623	1.5465

Framework versions

• Transformers 4.29.2
• Pytorch 2.0.1
• Datasets 2.12.0
• Tokenizers 0.13.3