distilbert-base-uncased-finetuned-ner-cadec-active

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.3506
Precision: 0.5141
Recall: 0.5609
F1: 0.5365
Accuracy: 0.9010
Adr Precision: 0.4552
Adr Recall: 0.5683
Adr F1: 0.5055
Disease Precision: 0.0
Disease Recall: 0.0
Disease F1: 0.0
Drug Precision: 0.7659
Drug Recall: 0.8351
Drug F1: 0.7990
Finding Precision: 0.0
Finding Recall: 0.0
Finding F1: 0.0
Symptom Precision: 0.0
Symptom Recall: 0.0
Symptom F1: 0.0
B-adr Precision: 0.6573
B-adr Recall: 0.7339
B-adr F1: 0.6935
B-disease Precision: 0.0
B-disease Recall: 0.0
B-disease F1: 0.0
B-drug Precision: 0.9318
B-drug Recall: 0.8723
B-drug F1: 0.9011
B-finding Precision: 0.0
B-finding Recall: 0.0
B-finding F1: 0.0
B-symptom Precision: 0.0
B-symptom Recall: 0.0
B-symptom F1: 0.0
I-adr Precision: 0.4521
I-adr Recall: 0.5422
I-adr F1: 0.4931
I-disease Precision: 0.0
I-disease Recall: 0.0
I-disease F1: 0.0
I-drug Precision: 0.7960
I-drug Recall: 0.8556
I-drug F1: 0.8247
I-finding Precision: 0.0
I-finding Recall: 0.0
I-finding F1: 0.0
I-symptom Precision: 0.0
I-symptom Recall: 0.0
I-symptom F1: 0.0
Macro Avg F1: 0.2912
Weighted Avg F1: 0.5966

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: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy	Adr Precision	Adr Recall	Adr F1	Drug Precision	Drug Recall	Drug F1	B-adr Precision	B-adr Recall	B-adr F1	B-drug Precision	B-drug Recall	B-drug F1	I-adr Precision	I-adr Recall	I-adr F1	I-drug Precision	I-drug Recall	I-drug F1	Macro Avg F1	Weighted Avg F1
No log	1.0	26	0.6925	0.0	0.0	0.0	0.7877	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
No log	2.0	52	0.4818	0.3775	0.2508	0.3014	0.8527	0.2356	0.1788	0.2033	0.9606	0.6489	0.7746	0.0	0.0	0.0	0.9841	0.6596	0.7898	0.0632	0.0592	0.0612	0.9685	0.6578	0.7834	0.1634	0.1890
No log	3.0	78	0.4090	0.3740	0.3234	0.3469	0.8761	0.2779	0.2747	0.2763	0.7697	0.6755	0.7195	0.6431	0.2583	0.3685	0.9524	0.7447	0.8358	0.1645	0.1813	0.1725	0.8375	0.7166	0.7723	0.2149	0.3628
No log	4.0	104	0.3834	0.4128	0.3562	0.3824	0.8813	0.3096	0.3038	0.3067	0.8274	0.7394	0.7809	0.6346	0.4047	0.4942	0.9664	0.7660	0.8546	0.2382	0.2442	0.2411	0.8735	0.7754	0.8215	0.2411	0.4379
No log	5.0	130	0.3587	0.4437	0.4882	0.4649	0.8960	0.3705	0.4782	0.4175	0.7914	0.7872	0.7893	0.6234	0.6803	0.6506	0.9682	0.8085	0.8812	0.3689	0.4345	0.3990	0.8065	0.8021	0.8043	0.2735	0.5465
No log	6.0	156	0.3549	0.4600	0.5005	0.4794	0.8960	0.3876	0.4913	0.4333	0.7906	0.8032	0.7968	0.6281	0.7102	0.6667	0.9625	0.8191	0.8851	0.3861	0.4596	0.4197	0.8211	0.8342	0.8276	0.2799	0.5619
No log	7.0	182	0.3583	0.4842	0.5333	0.5075	0.9013	0.4205	0.5305	0.4692	0.75	0.8298	0.7879	0.6412	0.7260	0.6809	0.9310	0.8617	0.8950	0.4238	0.4794	0.4499	0.7892	0.8610	0.8235	0.2849	0.5774
No log	8.0	208	0.3442	0.4954	0.5455	0.5192	0.8986	0.4320	0.5451	0.4820	0.7670	0.8404	0.8020	0.6561	0.7150	0.6843	0.9326	0.8830	0.9071	0.4279	0.5224	0.4705	0.8010	0.8610	0.8299	0.2892	0.5872
No log	9.0	234	0.3499	0.5075	0.5517	0.5287	0.9003	0.4480	0.5567	0.4964	0.7610	0.8298	0.7939	0.6585	0.7228	0.6892	0.9314	0.8670	0.8981	0.4466	0.5404	0.4890	0.7960	0.8556	0.8247	0.2901	0.5934
No log	10.0	260	0.3506	0.5141	0.5609	0.5365	0.9010	0.4552	0.5683	0.5055	0.7659	0.8351	0.7990	0.6573	0.7339	0.6935	0.9318	0.8723	0.9011	0.4521	0.5422	0.4931	0.7960	0.8556	0.8247	0.2912	0.5966

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.15.0
Tokenizers 0.15.0

csNoHug/distilbert-base-uncased-finetuned-ner-cadec-active

distilbert-base-uncased-finetuned-ner-cadec-active

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

为此模型做种