bert-base-uncased-finetunedINv-ner

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: 1.3930
• Precision: 0.5290
• Recall: 0.3557
• F1: 0.4254
• Accuracy: 0.9049

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: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 60

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	126	0.8572	0.2799	0.0422	0.0734	0.8882
No log	2.0	252	0.8837	0.4222	0.2194	0.2888	0.8942
No log	3.0	378	0.9128	0.4662	0.2663	0.3390	0.8974
0.2047	4.0	504	0.9186	0.5046	0.3176	0.3898	0.9011
0.2047	5.0	630	0.9562	0.5030	0.3253	0.3951	0.9022
0.2047	6.0	756	0.9832	0.5068	0.3385	0.4059	0.9030
0.2047	7.0	882	0.9915	0.5064	0.3583	0.4197	0.9043
0.0586	8.0	1008	1.0205	0.5236	0.3494	0.4191	0.9044
0.0586	9.0	1134	1.0542	0.5337	0.3448	0.4189	0.9043
0.0586	10.0	1260	1.0543	0.5257	0.3607	0.4278	0.9044
0.0586	11.0	1386	1.0989	0.5228	0.3585	0.4254	0.9044
0.0296	12.0	1512	1.1170	0.5290	0.3540	0.4241	0.9043
0.0296	13.0	1638	1.0868	0.5131	0.3665	0.4276	0.9047
0.0296	14.0	1764	1.1363	0.5276	0.3521	0.4224	0.9045
0.0296	15.0	1890	1.1289	0.4817	0.3736	0.4208	0.9045
0.0186	16.0	2016	1.1465	0.4978	0.3570	0.4158	0.9044
0.0186	17.0	2142	1.1430	0.5098	0.3637	0.4245	0.9046
0.0186	18.0	2268	1.1732	0.5176	0.3637	0.4272	0.9048
0.0186	19.0	2394	1.1804	0.5175	0.3593	0.4241	0.9047
0.0126	20.0	2520	1.1722	0.5244	0.3618	0.4282	0.9053
0.0126	21.0	2646	1.1898	0.5109	0.3590	0.4217	0.9045
0.0126	22.0	2772	1.2175	0.5275	0.3551	0.4245	0.9047
0.0126	23.0	2898	1.2256	0.5280	0.3521	0.4225	0.9046
0.009	24.0	3024	1.2582	0.5314	0.3465	0.4194	0.9043
0.009	25.0	3150	1.2385	0.5234	0.3578	0.4250	0.9045
0.009	26.0	3276	1.2306	0.5148	0.3646	0.4269	0.9049
0.009	27.0	3402	1.2540	0.5270	0.3543	0.4238	0.9048
0.0065	28.0	3528	1.2523	0.5138	0.3550	0.4199	0.9046
0.0065	29.0	3654	1.2830	0.5392	0.3552	0.4283	0.9048
0.0065	30.0	3780	1.2749	0.5215	0.3590	0.4252	0.9048
0.0065	31.0	3906	1.2931	0.5295	0.3564	0.4260	0.9047
0.0048	32.0	4032	1.3002	0.5320	0.3464	0.4196	0.9044
0.0048	33.0	4158	1.3068	0.5218	0.3584	0.4249	0.9046
0.0048	34.0	4284	1.3021	0.5180	0.3609	0.4254	0.9050
0.0048	35.0	4410	1.3129	0.5160	0.3583	0.4230	0.9047
0.0037	36.0	4536	1.3258	0.5366	0.3521	0.4252	0.9044
0.0037	37.0	4662	1.3178	0.5390	0.3564	0.4291	0.9048
0.0037	38.0	4788	1.3253	0.5307	0.3567	0.4267	0.9047
0.0037	39.0	4914	1.3231	0.5230	0.3584	0.4254	0.9049
0.0029	40.0	5040	1.3442	0.5372	0.3547	0.4273	0.9047
0.0029	41.0	5166	1.3441	0.5237	0.3537	0.4223	0.9046
0.0029	42.0	5292	1.3513	0.5359	0.3551	0.4272	0.9048
0.0029	43.0	5418	1.3616	0.5459	0.3526	0.4284	0.9048
0.0023	44.0	5544	1.3574	0.5335	0.3509	0.4233	0.9046
0.0023	45.0	5670	1.3469	0.5261	0.3599	0.4274	0.9050
0.0023	46.0	5796	1.3761	0.5456	0.3565	0.4312	0.9049
0.0023	47.0	5922	1.3596	0.5326	0.3609	0.4302	0.9050
0.0018	48.0	6048	1.3696	0.5323	0.3561	0.4268	0.9049
0.0018	49.0	6174	1.3751	0.5274	0.3553	0.4245	0.9049
0.0018	50.0	6300	1.3806	0.5295	0.3563	0.4259	0.9048
0.0018	51.0	6426	1.3787	0.5325	0.3579	0.4281	0.9050
0.0015	52.0	6552	1.3803	0.5270	0.3567	0.4254	0.9048
0.0015	53.0	6678	1.3779	0.5304	0.3588	0.4280	0.9051
0.0015	54.0	6804	1.3768	0.5232	0.3578	0.4249	0.9049
0.0015	55.0	6930	1.3854	0.5258	0.3568	0.4251	0.9049
0.0013	56.0	7056	1.3936	0.5313	0.3547	0.4254	0.9049
0.0013	57.0	7182	1.3894	0.5257	0.3572	0.4254	0.9049
0.0013	58.0	7308	1.3946	0.5303	0.3552	0.4254	0.9048
0.0013	59.0	7434	1.3922	0.5275	0.3560	0.4251	0.9049
0.0012	60.0	7560	1.3930	0.5290	0.3557	0.4254	0.9049

Framework versions

• Transformers 4.49.0
• Pytorch 2.7.1+cu118
• Datasets 4.7.0
• Tokenizers 0.21.4