metadata

license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
datasets:
  - ktgiahieu/maccrobat2018_2020
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: BioMedical_NER-maccrobat-distilbert
    results: []
widget:
  - text: >-
      CASE: A 28-year-old previously healthy man presented with a 6-week history
      of palpitations. The symptoms occurred during rest, 2–3 times per week,
      lasted up to 30 minutes at a time and were associated with dyspnea. Except
      for a grade 2/6 holosystolic tricuspid regurgitation murmur (best heard at
      the left sternal border with inspiratory accentuation), physical
      examination yielded unremarkable findings.
    example_title: example 1
  - text: >-
      A 63-year-old woman with no known cardiac history presented with a sudden
      onset of dyspnea requiring intubation and ventilatory support out of
      hospital. She denied preceding symptoms of chest discomfort, palpitations,
      syncope or infection. The patient was afebrile and normotensive, with a
      sinus tachycardia of 140 beats/min.
    example_title: example 2
  - text: >-
      A 48 year-old female presented with vaginal bleeding and abnormal Pap
      smears. Upon diagnosis of invasive non-keratinizing SCC of the cervix, she
      underwent a radical hysterectomy with salpingo-oophorectomy which
      demonstrated positive spread to the pelvic lymph nodes and the
      parametrium. Pathological examination revealed that the tumour also
      extensively involved the lower uterine segment.
    example_title: example 3

BioMedical_NER-maccrobat-distilbert

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

Loss: 0.3418
Precision: 0.8858
Recall: 0.9578
F1: 0.9204
Accuracy: 0.9541

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: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 70

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	45	1.8297	0.0	0.0	0.0	0.6197
No log	2.0	90	1.5738	0.2713	0.0490	0.0830	0.6324
No log	3.0	135	1.3283	0.3165	0.2269	0.2644	0.6654
No log	4.0	180	1.1738	0.3634	0.3538	0.3585	0.6915
No log	5.0	225	1.1003	0.4080	0.5041	0.4510	0.7074
No log	6.0	270	1.0484	0.4339	0.5727	0.4937	0.7193
No log	7.0	315	0.9841	0.4685	0.6209	0.5340	0.7434
No log	8.0	360	0.8765	0.5286	0.6369	0.5777	0.7712
No log	9.0	405	0.8037	0.5638	0.6635	0.6096	0.7922
No log	10.0	450	0.7924	0.5572	0.7013	0.6210	0.8008
No log	11.0	495	0.7403	0.5732	0.7228	0.6394	0.8143
1.0716	12.0	540	0.6235	0.6636	0.7083	0.6852	0.8457
1.0716	13.0	585	0.6182	0.6418	0.7448	0.6895	0.8487
1.0716	14.0	630	0.6498	0.6312	0.7724	0.6947	0.8456
1.0716	15.0	675	0.5830	0.6638	0.7874	0.7204	0.8650
1.0716	16.0	720	0.5199	0.6992	0.7954	0.7442	0.8804
1.0716	17.0	765	0.5470	0.7129	0.8119	0.7592	0.8836
1.0716	18.0	810	0.5065	0.7269	0.8318	0.7758	0.8920
1.0716	19.0	855	0.4645	0.7521	0.8353	0.7916	0.9018
1.0716	20.0	900	0.5204	0.7240	0.8501	0.7820	0.8915
1.0716	21.0	945	0.4383	0.7660	0.8495	0.8056	0.9078
1.0716	22.0	990	0.4345	0.7659	0.8662	0.8130	0.9127
0.2987	23.0	1035	0.4492	0.7675	0.8733	0.8170	0.9118
0.2987	24.0	1080	0.4654	0.7691	0.8805	0.8211	0.9101
0.2987	25.0	1125	0.4186	0.7995	0.8778	0.8368	0.9216
0.2987	26.0	1170	0.3898	0.8131	0.8871	0.8485	0.9269
0.2987	27.0	1215	0.4057	0.8041	0.8928	0.8461	0.9256
0.2987	28.0	1260	0.3916	0.8156	0.8938	0.8529	0.9290
0.2987	29.0	1305	0.3771	0.8250	0.8989	0.8604	0.9317
0.2987	30.0	1350	0.3690	0.8253	0.8997	0.8609	0.9337
0.2987	31.0	1395	0.3716	0.8320	0.9084	0.8685	0.9357
0.2987	32.0	1440	0.3764	0.8278	0.9115	0.8677	0.9349
0.2987	33.0	1485	0.3549	0.8389	0.9113	0.8736	0.9376
0.1133	34.0	1530	0.3715	0.8368	0.9160	0.8746	0.9372
0.1133	35.0	1575	0.3621	0.8452	0.9208	0.8814	0.9401
0.1133	36.0	1620	0.3533	0.8489	0.9248	0.8852	0.9420
0.1133	37.0	1665	0.3471	0.8540	0.9259	0.8885	0.9427
0.1133	38.0	1710	0.3492	0.8504	0.9263	0.8867	0.9423
0.1133	39.0	1755	0.3570	0.8572	0.9327	0.8933	0.9441
0.1133	40.0	1800	0.3647	0.8535	0.9348	0.8923	0.9436
0.1133	41.0	1845	0.3500	0.8656	0.9381	0.9004	0.9466
0.1133	42.0	1890	0.3570	0.8594	0.9405	0.8981	0.9452
0.1133	43.0	1935	0.3545	0.8695	0.9436	0.9050	0.9480
0.1133	44.0	1980	0.3578	0.8660	0.9415	0.9022	0.9467
0.0575	45.0	2025	0.3384	0.8723	0.9419	0.9058	0.9498
0.0575	46.0	2070	0.3450	0.8755	0.9472	0.9100	0.9502
0.0575	47.0	2115	0.3468	0.8736	0.9495	0.9100	0.9500
0.0575	48.0	2160	0.3488	0.8706	0.9502	0.9087	0.9505
0.0575	49.0	2205	0.3480	0.8738	0.9517	0.9111	0.9506
0.0575	50.0	2250	0.3474	0.8725	0.9504	0.9098	0.9501
0.0575	51.0	2295	0.3463	0.8711	0.9498	0.9087	0.9499
0.0575	52.0	2340	0.3328	0.8782	0.9525	0.9138	0.9518
0.0575	53.0	2385	0.3550	0.8738	0.9527	0.9115	0.9508
0.0575	54.0	2430	0.3351	0.8777	0.9525	0.9135	0.9526
0.0575	55.0	2475	0.3438	0.8781	0.9548	0.9148	0.9521
0.0364	56.0	2520	0.3452	0.8797	0.9540	0.9153	0.9521
0.0364	57.0	2565	0.3496	0.8810	0.9561	0.9170	0.9523
0.0364	58.0	2610	0.3472	0.8802	0.9557	0.9164	0.9525
0.0364	59.0	2655	0.3476	0.8813	0.9559	0.9171	0.9530
0.0364	60.0	2700	0.3413	0.8839	0.9563	0.9187	0.9536
0.0364	61.0	2745	0.3395	0.8839	0.9563	0.9187	0.9538
0.0364	62.0	2790	0.3417	0.8843	0.9580	0.9196	0.9537
0.0364	63.0	2835	0.3397	0.8846	0.9563	0.9191	0.9536
0.0364	64.0	2880	0.3428	0.8839	0.9576	0.9192	0.9534
0.0364	65.0	2925	0.3411	0.8847	0.9576	0.9197	0.9539
0.0364	66.0	2970	0.3442	0.8849	0.9574	0.9197	0.9538
0.028	67.0	3015	0.3444	0.8844	0.9578	0.9196	0.9538
0.028	68.0	3060	0.3437	0.8857	0.9584	0.9206	0.9541
0.028	69.0	3105	0.3411	0.8857	0.9582	0.9205	0.9540
0.028	70.0	3150	0.3418	0.8858	0.9578	0.9204	0.9541

Framework versions

Transformers 4.32.1
Pytorch 2.0.1+cu118
Datasets 2.14.4
Tokenizers 0.13.3