distilbert-base-multilingual-NER-naamapdam-fine-tuned

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

• Loss: 0.3852
• Precision: 0.7940
• Recall: 0.8182
• F1: 0.8059
• Accuracy: 0.9124

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: 256
• eval_batch_size: 512
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.3757	0.51	1000	0.3173	0.7666	0.7858	0.7761	0.8984
0.3062	1.03	2000	0.3020	0.7791	0.7981	0.7885	0.9026
0.2793	1.54	3000	0.2962	0.7827	0.8021	0.7923	0.9059
0.2755	2.06	4000	0.2973	0.7768	0.8122	0.7941	0.9048
0.2529	2.57	5000	0.2879	0.7747	0.8201	0.7968	0.9057
0.2483	3.08	6000	0.3025	0.7714	0.8298	0.7996	0.9079
0.2294	3.6	7000	0.2899	0.7877	0.8211	0.8041	0.9105
0.2252	4.11	8000	0.2952	0.7850	0.8185	0.8014	0.9090
0.2088	4.63	9000	0.2932	0.7851	0.8234	0.8038	0.9090
0.2046	5.14	10000	0.2998	0.7931	0.8215	0.8071	0.9117
0.1909	5.66	11000	0.3029	0.7925	0.8240	0.8080	0.9112
0.1857	6.17	12000	0.3160	0.7903	0.8228	0.8062	0.9108
0.1744	6.68	13000	0.3099	0.7858	0.8259	0.8054	0.9115
0.1686	7.2	14000	0.3199	0.7859	0.8246	0.8048	0.9097
0.1613	7.71	15000	0.3161	0.7941	0.8179	0.8058	0.9121
0.1538	8.23	16000	0.3294	0.7903	0.8221	0.8059	0.9110
0.1475	8.74	17000	0.3260	0.7935	0.8248	0.8089	0.9129
0.1429	9.25	18000	0.3378	0.7958	0.8210	0.8082	0.9130
0.1369	9.77	19000	0.3402	0.7905	0.8240	0.8069	0.9118
0.1302	10.28	20000	0.3573	0.7865	0.8269	0.8062	0.9114
0.1276	10.8	21000	0.3564	0.7924	0.8208	0.8063	0.9117
0.122	11.31	22000	0.3590	0.7939	0.8274	0.8103	0.9130
0.1181	11.83	23000	0.3660	0.7974	0.8234	0.8102	0.9132
0.1141	12.34	24000	0.3695	0.7921	0.8208	0.8062	0.9112
0.1118	12.85	25000	0.3649	0.7942	0.8188	0.8063	0.9114
0.1081	13.37	26000	0.3781	0.7980	0.8149	0.8064	0.9124
0.1054	13.88	27000	0.3800	0.7913	0.8179	0.8044	0.9120
0.1023	14.4	28000	0.3857	0.7942	0.8207	0.8072	0.9128
0.101	14.91	29000	0.3852	0.7940	0.8182	0.8059	0.9124

Framework versions

• Transformers 4.27.4
• Pytorch 2.0.0+cu117
• Datasets 2.11.0
• Tokenizers 0.13.3