DIALOGUE

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: 0.0320
• Accuracy: 0.9902

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: 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: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.2742	0.31	15	1.0896	0.7353
0.9231	0.62	30	0.7436	0.8039
0.7035	0.94	45	0.4280	0.9706
0.4361	1.25	60	0.2308	1.0
0.3081	1.56	75	0.1590	0.9902
0.1794	1.88	90	0.1269	0.9706
0.0988	2.19	105	0.0605	0.9902
0.0606	2.5	120	0.0617	0.9902
0.0372	2.81	135	0.0474	0.9902
0.0209	3.12	150	0.0411	0.9902
0.0134	3.44	165	0.0326	0.9902
0.0099	3.75	180	0.0273	0.9902
0.0085	4.06	195	0.0348	0.9902
0.0065	4.38	210	0.0335	0.9902
0.0058	4.69	225	0.0318	0.9902
0.0047	5.0	240	0.0309	0.9902
0.0041	5.31	255	0.0289	0.9902
0.0041	5.62	270	0.0284	0.9902
0.0038	5.94	285	0.0275	0.9902
0.0036	6.25	300	0.0270	0.9902
0.003	6.56	315	0.0260	0.9902
0.0027	6.88	330	0.0270	0.9902
0.0026	7.19	345	0.0291	0.9902
0.0024	7.5	360	0.0298	0.9902
0.0023	7.81	375	0.0283	0.9902
0.0024	8.12	390	0.0255	0.9902
0.0021	8.44	405	0.0251	0.9902
0.0019	8.75	420	0.0260	0.9902
0.0019	9.06	435	0.0264	0.9902
0.0017	9.38	450	0.0266	0.9902
0.0017	9.69	465	0.0276	0.9902
0.0017	10.0	480	0.0271	0.9902
0.0015	10.31	495	0.0271	0.9902
0.0015	10.62	510	0.0271	0.9902
0.0015	10.94	525	0.0266	0.9902
0.0014	11.25	540	0.0261	0.9902
0.0013	11.56	555	0.0260	0.9902
0.0013	11.88	570	0.0264	0.9902
0.0013	12.19	585	0.0268	0.9902
0.0013	12.5	600	0.0270	0.9902
0.0012	12.81	615	0.0274	0.9902
0.0012	13.12	630	0.0274	0.9902
0.0012	13.44	645	0.0278	0.9902
0.0011	13.75	660	0.0278	0.9902
0.001	14.06	675	0.0278	0.9902
0.001	14.38	690	0.0283	0.9902
0.001	14.69	705	0.0290	0.9902
0.001	15.0	720	0.0286	0.9902
0.001	15.31	735	0.0284	0.9902
0.001	15.62	750	0.0285	0.9902
0.0009	15.94	765	0.0289	0.9902
0.0009	16.25	780	0.0298	0.9902
0.0009	16.56	795	0.0305	0.9902
0.0009	16.88	810	0.0309	0.9902
0.0009	17.19	825	0.0304	0.9902
0.0008	17.5	840	0.0303	0.9902
0.0008	17.81	855	0.0302	0.9902
0.0009	18.12	870	0.0301	0.9902
0.0008	18.44	885	0.0300	0.9902
0.0008	18.75	900	0.0302	0.9902
0.0008	19.06	915	0.0300	0.9902
0.0007	19.38	930	0.0301	0.9902
0.0007	19.69	945	0.0299	0.9902
0.0008	20.0	960	0.0304	0.9902
0.0007	20.31	975	0.0302	0.9902
0.0007	20.62	990	0.0304	0.9902
0.0007	20.94	1005	0.0305	0.9902
0.0007	21.25	1020	0.0312	0.9902
0.0007	21.56	1035	0.0311	0.9902
0.0007	21.88	1050	0.0310	0.9902
0.0007	22.19	1065	0.0310	0.9902
0.0007	22.5	1080	0.0309	0.9902
0.0006	22.81	1095	0.0311	0.9902
0.0006	23.12	1110	0.0313	0.9902
0.0007	23.44	1125	0.0313	0.9902
0.0007	23.75	1140	0.0313	0.9902
0.0006	24.06	1155	0.0312	0.9902
0.0006	24.38	1170	0.0313	0.9902
0.0007	24.69	1185	0.0314	0.9902
0.0006	25.0	1200	0.0313	0.9902
0.0006	25.31	1215	0.0314	0.9902
0.0006	25.62	1230	0.0315	0.9902
0.0006	25.94	1245	0.0316	0.9902
0.0006	26.25	1260	0.0316	0.9902
0.0006	26.56	1275	0.0316	0.9902
0.0006	26.88	1290	0.0316	0.9902
0.0006	27.19	1305	0.0315	0.9902
0.0006	27.5	1320	0.0319	0.9902
0.0006	27.81	1335	0.0320	0.9902
0.0006	28.12	1350	0.0320	0.9902
0.0006	28.44	1365	0.0320	0.9902
0.0006	28.75	1380	0.0320	0.9902
0.0006	29.06	1395	0.0320	0.9902
0.0006	29.38	1410	0.0320	0.9902
0.0006	29.69	1425	0.0320	0.9902
0.0006	30.0	1440	0.0320	0.9902

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0