distilbert-base-uncased-finetuned-ner

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: 0.4331
• Precision: 0.7801
• Recall: 0.7129
• F1: 0.745
• Accuracy: 0.9363

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: 16
• eval_batch_size: 16
• 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	Precision	Recall	F1	Accuracy
No log	1.0	23	0.8283	0.0	0.0	0.0	0.8332
No log	2.0	46	0.5781	0.4474	0.0813	0.1377	0.8482
No log	3.0	69	0.4927	0.4841	0.3636	0.4153	0.8913
No log	4.0	92	0.4348	0.4657	0.4545	0.4600	0.9044
No log	5.0	115	0.4293	0.5561	0.4976	0.5253	0.9088
No log	6.0	138	0.3934	0.6313	0.5981	0.6143	0.9244
No log	7.0	161	0.3961	0.7219	0.6459	0.6818	0.9313
No log	8.0	184	0.3648	0.7098	0.6555	0.6816	0.9325
No log	9.0	207	0.3961	0.7582	0.6603	0.7059	0.9357
No log	10.0	230	0.3800	0.7474	0.6794	0.7118	0.9350
No log	11.0	253	0.3661	0.7474	0.6794	0.7118	0.9332
No log	12.0	276	0.3697	0.7619	0.6890	0.7236	0.9344
No log	13.0	299	0.3829	0.7660	0.6890	0.7254	0.9350
No log	14.0	322	0.3859	0.7849	0.6986	0.7392	0.9350
No log	15.0	345	0.3760	0.7946	0.7033	0.7462	0.9375
No log	16.0	368	0.3609	0.7602	0.7129	0.7358	0.9357
No log	17.0	391	0.3687	0.7766	0.6986	0.7355	0.9350
No log	18.0	414	0.3856	0.8043	0.7081	0.7532	0.9375
No log	19.0	437	0.3901	0.7861	0.7033	0.7424	0.9369
No log	20.0	460	0.4151	0.8276	0.6890	0.7520	0.9388
No log	21.0	483	0.3892	0.7824	0.7225	0.7512	0.9382
0.1775	22.0	506	0.3952	0.7947	0.7225	0.7569	0.9375
0.1775	23.0	529	0.3906	0.7817	0.7368	0.7586	0.9382
0.1775	24.0	552	0.4132	0.8156	0.6986	0.7526	0.9413
0.1775	25.0	575	0.4048	0.7979	0.7177	0.7557	0.9388
0.1775	26.0	598	0.4026	0.7772	0.7177	0.7463	0.9363
0.1775	27.0	621	0.4084	0.7789	0.7081	0.7419	0.9363
0.1775	28.0	644	0.4081	0.7865	0.7225	0.7531	0.9375
0.1775	29.0	667	0.4058	0.7795	0.7273	0.7525	0.9375
0.1775	30.0	690	0.4100	0.7772	0.7177	0.7463	0.9369
0.1775	31.0	713	0.4146	0.7824	0.7225	0.7512	0.9363
0.1775	32.0	736	0.4142	0.7865	0.7225	0.7531	0.9363
0.1775	33.0	759	0.4168	0.7824	0.7225	0.7512	0.9369
0.1775	34.0	782	0.4367	0.8122	0.7033	0.7538	0.9388
0.1775	35.0	805	0.4282	0.7884	0.7129	0.7487	0.9363
0.1775	36.0	828	0.4249	0.7842	0.7129	0.7469	0.9357
0.1775	37.0	851	0.4297	0.7884	0.7129	0.7487	0.9363
0.1775	38.0	874	0.4218	0.7824	0.7225	0.7512	0.9375
0.1775	39.0	897	0.4267	0.7842	0.7129	0.7469	0.9363
0.1775	40.0	920	0.4272	0.7937	0.7177	0.7538	0.9369
0.1775	41.0	943	0.4308	0.7926	0.7129	0.7506	0.9363
0.1775	42.0	966	0.4390	0.7884	0.7129	0.7487	0.9363
0.1775	43.0	989	0.4366	0.7914	0.7081	0.7475	0.9375
0.0065	44.0	1012	0.4311	0.7749	0.7081	0.74	0.9350
0.0065	45.0	1035	0.4276	0.7760	0.7129	0.7431	0.9357
0.0065	46.0	1058	0.4313	0.7801	0.7129	0.745	0.9357
0.0065	47.0	1081	0.4330	0.7801	0.7129	0.745	0.9357
0.0065	48.0	1104	0.4325	0.7801	0.7129	0.745	0.9363
0.0065	49.0	1127	0.4328	0.7801	0.7129	0.745	0.9363
0.0065	50.0	1150	0.4331	0.7801	0.7129	0.745	0.9363

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.2+cu118
• Datasets 2.16.0
• Tokenizers 0.15.0