distilbert-base-uncased-finetuned-IAM

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

• Loss: 0.9814
• Accuracy: 0.5103
• F1: 0.4950

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: 10
• eval_batch_size: 10
• 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	F1
1.5871	1.0	15	1.4971	0.3379	0.1821
1.4995	2.0	30	1.4588	0.3379	0.1707
1.464	3.0	45	1.4251	0.3655	0.2870
1.4105	4.0	60	1.4027	0.3793	0.2899
1.4269	5.0	75	1.3798	0.3793	0.2899
1.3835	6.0	90	1.3425	0.3724	0.3087
1.3885	7.0	105	1.3041	0.4069	0.3515
1.3286	8.0	120	1.3004	0.4621	0.4450
1.3572	9.0	135	1.2621	0.4345	0.3903
1.3176	10.0	150	1.2033	0.4552	0.4250
1.2509	11.0	165	1.1942	0.5034	0.4755
1.2781	12.0	180	1.1689	0.4828	0.4651
1.2156	13.0	195	1.1438	0.5034	0.4837
1.1518	14.0	210	1.1187	0.5034	0.4844
1.161	15.0	225	1.1013	0.5034	0.4858
1.1377	16.0	240	1.0882	0.5034	0.4796
1.1634	17.0	255	1.0692	0.5034	0.4860
1.0666	18.0	270	1.0591	0.5034	0.4772
1.1358	19.0	285	1.0455	0.5034	0.4736
1.1118	20.0	300	1.0313	0.5034	0.4872
1.0367	21.0	315	1.0228	0.5034	0.4853
1.0781	22.0	330	1.0106	0.5034	0.4857
1.0346	23.0	345	1.0034	0.5034	0.4935
1.1015	24.0	360	1.0032	0.5034	0.4806
1.0147	25.0	375	0.9911	0.5103	0.4903
1.0144	26.0	390	0.9856	0.5103	0.4972
1.022	27.0	405	0.9835	0.5103	0.4982
1.0218	28.0	420	0.9821	0.5103	0.4955
1.0173	29.0	435	0.9811	0.5103	0.4950
1.0241	30.0	450	0.9814	0.5103	0.4950

Framework versions

• Transformers 4.24.0
• Pytorch 2.0.0
• Datasets 2.10.1
• Tokenizers 0.11.0