metadata

license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-medical-ner
    results: []

bert-medical-ner

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

Loss: 1.0318
Precision: 0.6083
Recall: 0.6344
F1: 0.6210
Accuracy: 0.7635

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: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	63	1.6731	0.3451	0.3549	0.3499	0.6003
No log	2.0	126	1.2481	0.4832	0.5248	0.5032	0.6912
No log	3.0	189	1.0959	0.5280	0.5703	0.5483	0.7198
No log	4.0	252	1.0258	0.5577	0.5878	0.5723	0.7330
No log	5.0	315	0.9761	0.5788	0.6038	0.5910	0.7433
No log	6.0	378	0.9461	0.5909	0.6068	0.5988	0.7478
No log	7.0	441	0.9456	0.5918	0.6143	0.6029	0.7516
1.1189	8.0	504	0.9396	0.5991	0.6164	0.6076	0.7562
1.1189	9.0	567	0.9594	0.6020	0.6252	0.6134	0.7569
1.1189	10.0	630	0.9742	0.6005	0.6203	0.6102	0.7555
1.1189	11.0	693	0.9700	0.6063	0.6256	0.6158	0.7597
1.1189	12.0	756	0.9772	0.5999	0.6246	0.6120	0.7582
1.1189	13.0	819	0.9890	0.6023	0.6254	0.6137	0.7593
1.1189	14.0	882	1.0011	0.6077	0.6332	0.6202	0.7631
1.1189	15.0	945	1.0096	0.6057	0.6305	0.6178	0.7613
0.4346	16.0	1008	1.0167	0.6112	0.6281	0.6195	0.7625
0.4346	17.0	1071	1.0227	0.6126	0.6357	0.6240	0.7651
0.4346	18.0	1134	1.0290	0.6072	0.6357	0.6212	0.7635
0.4346	19.0	1197	1.0288	0.6094	0.6363	0.6226	0.7643
0.4346	20.0	1260	1.0318	0.6083	0.6344	0.6210	0.7635

Framework versions

Transformers 4.29.2
Pytorch 2.0.1+cu117
Datasets 2.12.0
Tokenizers 0.13.3