bert-finetuned-ner

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

• Loss: 0.0274
• Precision: 0.9550
• Recall: 0.9638
• F1: 0.9594
• Accuracy: 0.9973

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use 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: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	148	0.0305	0.8341	0.8789	0.8559	0.9934
No log	2.0	296	0.0215	0.8834	0.9355	0.9087	0.9953
No log	3.0	444	0.0195	0.9140	0.9435	0.9285	0.9961
0.0655	4.0	592	0.0195	0.9282	0.9498	0.9389	0.9964
0.0655	5.0	740	0.0203	0.9177	0.9539	0.9355	0.9962
0.0655	6.0	888	0.0201	0.9401	0.9552	0.9475	0.9966
0.0056	7.0	1036	0.0200	0.9355	0.9535	0.9444	0.9968
0.0056	8.0	1184	0.0208	0.9393	0.9569	0.9480	0.9967
0.0056	9.0	1332	0.0215	0.9380	0.9549	0.9464	0.9968
0.0056	10.0	1480	0.0232	0.9188	0.9582	0.9381	0.9960
0.0024	11.0	1628	0.0212	0.9334	0.9554	0.9442	0.9967
0.0024	12.0	1776	0.0223	0.9383	0.9598	0.9489	0.9968
0.0024	13.0	1924	0.0225	0.9394	0.9542	0.9468	0.9967
0.0012	14.0	2072	0.0232	0.9415	0.9560	0.9487	0.9968
0.0012	15.0	2220	0.0238	0.9413	0.9580	0.9496	0.9967
0.0012	16.0	2368	0.0239	0.9396	0.9582	0.9488	0.9966
0.001	17.0	2516	0.0230	0.9328	0.9563	0.9444	0.9966
0.001	18.0	2664	0.0243	0.9342	0.9577	0.9458	0.9966
0.001	19.0	2812	0.0246	0.9423	0.9576	0.9499	0.9969
0.001	20.0	2960	0.0240	0.9355	0.9576	0.9464	0.9967
0.0006	21.0	3108	0.0241	0.9477	0.9599	0.9538	0.9970
0.0006	22.0	3256	0.0236	0.9443	0.9569	0.9505	0.9968
0.0006	23.0	3404	0.0244	0.9461	0.9578	0.9519	0.9969
0.0006	24.0	3552	0.0248	0.9417	0.96	0.9508	0.9969
0.0006	25.0	3700	0.0246	0.9336	0.9590	0.9461	0.9966
0.0006	26.0	3848	0.0236	0.9421	0.9589	0.9504	0.9968
0.0006	27.0	3996	0.0244	0.9441	0.9612	0.9526	0.9969
0.0004	28.0	4144	0.0250	0.9462	0.9594	0.9528	0.9969
0.0004	29.0	4292	0.0249	0.9430	0.9622	0.9525	0.9969
0.0004	30.0	4440	0.0252	0.9439	0.9612	0.9525	0.9969
0.0003	31.0	4588	0.0253	0.9480	0.9552	0.9515	0.9968
0.0003	32.0	4736	0.0229	0.9484	0.9619	0.9551	0.9969
0.0003	33.0	4884	0.0235	0.9485	0.9608	0.9546	0.9970
0.0003	34.0	5032	0.0247	0.9438	0.9611	0.9524	0.9969
0.0003	35.0	5180	0.0248	0.9481	0.9598	0.9539	0.9970
0.0003	36.0	5328	0.0245	0.9441	0.9621	0.9530	0.9969
0.0003	37.0	5476	0.0255	0.9417	0.9602	0.9508	0.9967
0.0002	38.0	5624	0.0255	0.9416	0.9595	0.9505	0.9969
0.0002	39.0	5772	0.0246	0.9524	0.9611	0.9567	0.9971
0.0002	40.0	5920	0.0254	0.9435	0.9611	0.9522	0.9969
0.0003	41.0	6068	0.0252	0.9386	0.9608	0.9496	0.9966
0.0003	42.0	6216	0.0257	0.9385	0.9601	0.9492	0.9968
0.0003	43.0	6364	0.0251	0.9491	0.9591	0.9541	0.9970
0.0002	44.0	6512	0.0251	0.9448	0.9610	0.9528	0.9970
0.0002	45.0	6660	0.0252	0.9508	0.9622	0.9565	0.9972
0.0002	46.0	6808	0.0252	0.9486	0.9613	0.9549	0.9971
0.0002	47.0	6956	0.0262	0.9498	0.9618	0.9558	0.9971
0.0001	48.0	7104	0.0263	0.9520	0.9624	0.9572	0.9971
0.0001	49.0	7252	0.0263	0.9521	0.9624	0.9573	0.9971
0.0001	50.0	7400	0.0260	0.9526	0.9618	0.9572	0.9972
0.0001	51.0	7548	0.0248	0.9493	0.9634	0.9563	0.9971
0.0001	52.0	7696	0.0255	0.9502	0.9618	0.9560	0.9971
0.0001	53.0	7844	0.0258	0.9522	0.9617	0.9569	0.9972
0.0001	54.0	7992	0.0258	0.9481	0.9615	0.9548	0.9970
0.0001	55.0	8140	0.0251	0.9520	0.9617	0.9568	0.9972
0.0001	56.0	8288	0.0250	0.9509	0.9608	0.9558	0.9972
0.0001	57.0	8436	0.0260	0.9462	0.9601	0.9531	0.9972
0.0001	58.0	8584	0.0252	0.9563	0.9628	0.9595	0.9973
0.0001	59.0	8732	0.0247	0.9506	0.9624	0.9565	0.9972
0.0001	60.0	8880	0.0251	0.9510	0.9611	0.9560	0.9972
0.0001	61.0	9028	0.0255	0.9495	0.9614	0.9554	0.9972
0.0001	62.0	9176	0.0259	0.9537	0.9613	0.9575	0.9972
0.0001	63.0	9324	0.0259	0.9506	0.9609	0.9557	0.9972
0.0001	64.0	9472	0.0260	0.9544	0.9595	0.9569	0.9972
0.0	65.0	9620	0.0253	0.9511	0.9604	0.9557	0.9972
0.0	66.0	9768	0.0257	0.9526	0.9604	0.9565	0.9972
0.0	67.0	9916	0.0263	0.9528	0.9605	0.9566	0.9972
0.0	68.0	10064	0.0271	0.9544	0.9598	0.9571	0.9972
0.0	69.0	10212	0.0269	0.9530	0.9611	0.9571	0.9972
0.0	70.0	10360	0.0273	0.9514	0.9609	0.9561	0.9972
0.0	71.0	10508	0.0275	0.9535	0.9612	0.9573	0.9972
0.0	72.0	10656	0.0275	0.9524	0.9632	0.9578	0.9972
0.0	73.0	10804	0.0279	0.9537	0.9596	0.9566	0.9972
0.0	74.0	10952	0.0277	0.9475	0.9633	0.9554	0.9970
0.0	75.0	11100	0.0272	0.9537	0.9614	0.9575	0.9972
0.0	76.0	11248	0.0269	0.9541	0.9619	0.9580	0.9972
0.0	77.0	11396	0.0271	0.9552	0.9625	0.9588	0.9972
0.0	78.0	11544	0.0274	0.9457	0.9619	0.9537	0.9970
0.0	79.0	11692	0.0273	0.9524	0.9616	0.9570	0.9972
0.0	80.0	11840	0.0275	0.9530	0.9632	0.9581	0.9972
0.0	81.0	11988	0.0271	0.9496	0.9639	0.9567	0.9971
0.0	82.0	12136	0.0280	0.9537	0.9614	0.9575	0.9972
0.0	83.0	12284	0.0277	0.9499	0.9642	0.9570	0.9970
0.0	84.0	12432	0.0275	0.9517	0.9621	0.9569	0.9971
0.0	85.0	12580	0.0277	0.9524	0.9635	0.9579	0.9972
0.0	86.0	12728	0.0275	0.9517	0.9648	0.9582	0.9972
0.0	87.0	12876	0.0276	0.9519	0.9636	0.9577	0.9972
0.0	88.0	13024	0.0276	0.9541	0.9647	0.9594	0.9972
0.0	89.0	13172	0.0275	0.9500	0.9642	0.9571	0.9971
0.0	90.0	13320	0.0276	0.9532	0.9635	0.9584	0.9972
0.0	91.0	13468	0.0273	0.9542	0.9636	0.9589	0.9972
0.0	92.0	13616	0.0274	0.9541	0.9636	0.9588	0.9973
0.0	93.0	13764	0.0274	0.9552	0.9638	0.9595	0.9973
0.0	94.0	13912	0.0275	0.9547	0.9636	0.9591	0.9973
0.0	95.0	14060	0.0274	0.9557	0.9639	0.9598	0.9973
0.0	96.0	14208	0.0274	0.9548	0.9638	0.9593	0.9973
0.0	97.0	14356	0.0274	0.9550	0.9641	0.9595	0.9973
0.0	98.0	14504	0.0275	0.9552	0.9643	0.9597	0.9973
0.0	99.0	14652	0.0274	0.9549	0.9638	0.9593	0.9973
0.0	100.0	14800	0.0274	0.9550	0.9638	0.9594	0.9973

Framework versions

• Transformers 4.46.3
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3