test-ner

This model is a fine-tuned version of bert-base-uncased on ktgiahieu/maccrobat2018_2020 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3155
• Precision: 0.9157
• Recall: 0.9323
• F1: 0.9239
• Accuracy: 0.9530

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: 4
• eval_batch_size: 4
• 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
0.0316	1.25	100	0.3050	0.8799	0.9250	0.9019	0.9466
0.0296	2.5	200	0.2946	0.8904	0.9242	0.9070	0.9497
0.0258	3.75	300	0.2883	0.9006	0.9196	0.9100	0.9496
0.0227	5.0	400	0.2905	0.8888	0.9279	0.9079	0.9496
0.0183	6.25	500	0.2950	0.8864	0.9319	0.9086	0.9496
0.0161	7.5	600	0.2931	0.8914	0.9273	0.9090	0.9503
0.0153	8.75	700	0.3069	0.8947	0.9309	0.9124	0.9502
0.0131	10.0	800	0.3032	0.8927	0.9261	0.9091	0.9497
0.0117	11.25	900	0.2935	0.9035	0.9295	0.9163	0.9526
0.0111	12.5	1000	0.3117	0.8993	0.9290	0.9139	0.9509
0.0089	13.75	1100	0.3247	0.8902	0.9293	0.9094	0.9502
0.0115	15.0	1200	0.3121	0.8875	0.9383	0.9122	0.9491
0.0088	16.25	1300	0.3038	0.9058	0.9301	0.9178	0.9523
0.0083	17.5	1400	0.3237	0.8928	0.9336	0.9127	0.9513
0.0095	18.75	1500	0.3223	0.8885	0.9389	0.9130	0.9493
0.0078	20.0	1600	0.3335	0.8898	0.9380	0.9133	0.9505
0.0088	21.25	1700	0.3004	0.9070	0.9332	0.9199	0.9534
0.0068	22.5	1800	0.3424	0.8913	0.9350	0.9126	0.9497
0.0065	23.75	1900	0.3150	0.9034	0.9338	0.9184	0.9538
0.011	25.0	2000	0.3097	0.9044	0.9341	0.9190	0.9523
0.0082	26.25	2100	0.3101	0.9057	0.9301	0.9177	0.9527
0.0108	27.5	2200	0.3143	0.9083	0.9311	0.9196	0.9525
0.0081	28.75	2300	0.3211	0.9011	0.9371	0.9188	0.9525
0.0089	30.0	2400	0.3357	0.8996	0.9362	0.9175	0.9509
0.0074	31.25	2500	0.3097	0.9079	0.9305	0.9190	0.9517
0.0077	32.5	2600	0.3253	0.9032	0.9373	0.9199	0.9511
0.0076	33.75	2700	0.3252	0.9056	0.9337	0.9194	0.9526
0.0058	35.0	2800	0.3422	0.8981	0.9382	0.9177	0.9512
0.0067	36.25	2900	0.3323	0.9074	0.9375	0.9222	0.9532
0.0063	37.5	3000	0.3390	0.9066	0.9342	0.9202	0.9521
0.0053	38.75	3100	0.3241	0.9095	0.9374	0.9232	0.9537
0.0054	40.0	3200	0.3211	0.9017	0.9401	0.9205	0.9534
0.0051	41.25	3300	0.3339	0.8931	0.9407	0.9163	0.9500
0.0064	42.5	3400	0.3514	0.8977	0.9373	0.9170	0.9517
0.0056	43.75	3500	0.3327	0.9069	0.9371	0.9218	0.9528
0.0053	45.0	3600	0.3344	0.9034	0.9356	0.9192	0.9525
0.0048	46.25	3700	0.3203	0.9171	0.9355	0.9262	0.9542
0.0063	47.5	3800	0.3293	0.9109	0.9364	0.9234	0.9530
0.0037	48.75	3900	0.3375	0.9146	0.9315	0.9230	0.9520
0.0056	50.0	4000	0.3155	0.9157	0.9323	0.9239	0.9530

Ner Labels

"O", "B-ACTIVITY", "I-ACTIVITY", "I-ADMINISTRATION", "B-ADMINISTRATION", "B-AGE", "I-AGE", "I-AREA", "B-AREA", "B-BIOLOGICAL_ATTRIBUTE", "I-BIOLOGICAL_ATTRIBUTE", "I-BIOLOGICAL_STRUCTURE", "B-BIOLOGICAL_STRUCTURE", "B-CLINICAL_EVENT", "I-CLINICAL_EVENT", "B-COLOR", "I-COLOR", "I-COREFERENCE", "B-COREFERENCE", "B-DATE", "I-DATE", "I-DETAILED_DESCRIPTION", "B-DETAILED_DESCRIPTION", "I-DIAGNOSTIC_PROCEDURE", "B-DIAGNOSTIC_PROCEDURE", "I-DISEASE_DISORDER", "B-DISEASE_DISORDER", "B-DISTANCE", "I-DISTANCE", "B-DOSAGE", "I-DOSAGE", "I-DURATION", "B-DURATION", "I-FAMILY_HISTORY", "B-FAMILY_HISTORY", "B-FREQUENCY", "I-FREQUENCY", "I-HEIGHT", "B-HEIGHT", "B-HISTORY", "I-HISTORY", "I-LAB_VALUE", "B-LAB_VALUE", "I-MASS", "B-MASS", "I-MEDICATION", "B-MEDICATION", "I-NONBIOLOGICAL_LOCATION", "B-NONBIOLOGICAL_LOCATION", "I-OCCUPATION", "B-OCCUPATION", "B-OTHER_ENTITY", "I-OTHER_ENTITY", "B-OTHER_EVENT", "I-OTHER_EVENT", "I-OUTCOME", "B-OUTCOME", "I-PERSONAL_BACKGROUND", "B-PERSONAL_BACKGROUND", "B-QUALITATIVE_CONCEPT", "I-QUALITATIVE_CONCEPT", "I-QUANTITATIVE_CONCEPT", "B-QUANTITATIVE_CONCEPT", "B-SEVERITY", "I-SEVERITY", "B-SEX", "I-SEX", "B-SHAPE", "I-SHAPE", "B-SIGN_SYMPTOM", "I-SIGN_SYMPTOM", "B-SUBJECT", "I-SUBJECT", "B-TEXTURE", "I-TEXTURE", "B-THERAPEUTIC_PROCEDURE", "I-THERAPEUTIC_PROCEDURE", "I-TIME", "B-TIME", "B-VOLUME", "I-VOLUME", "I-WEIGHT", "B-WEIGHT",

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2