metadata

language:
  - en
license: mit
base_model: xlnet-base-cased
tags:
  - low-resource NER
  - token_classification
  - biomedicine
  - medical NER
  - generated_from_trainer
datasets:
  - medicine
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: Dagobert42/xlnet-base-cased-biored-augmented
    results: []

Dagobert42/xlnet-base-cased-biored-augmented

This model is a fine-tuned version of xlnet-base-cased on the bigbio/biored dataset. It achieves the following results on the evaluation set:

Loss: 0.1552
Accuracy: 0.9545
Precision: 0.8651
Recall: 0.8306
F1: 0.8454
Weighted F1: 0.9544

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: 1.8e-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
lr_scheduler_warmup_ratio: 0.004
num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Weighted F1
No log	0.5	10	0.2276	0.9252	0.7871	0.7482	0.7616	0.9233
No log	1.0	20	0.2124	0.9318	0.8363	0.7571	0.7923	0.9298
No log	1.5	30	0.2052	0.9342	0.8199	0.794	0.8057	0.9334
No log	2.0	40	0.1958	0.9396	0.8132	0.8049	0.8038	0.9384
No log	2.5	50	0.2043	0.9385	0.8162	0.8086	0.811	0.9377
No log	3.0	60	0.1948	0.9409	0.8413	0.8109	0.8249	0.9404
No log	3.5	70	0.1951	0.9436	0.8449	0.7963	0.8186	0.9425
No log	4.0	80	0.2032	0.941	0.8169	0.8158	0.8158	0.9411
No log	4.5	90	0.1984	0.944	0.827	0.8125	0.8194	0.9435
No log	5.0	100	0.1982	0.9451	0.8313	0.8072	0.8184	0.9443
No log	5.5	110	0.1968	0.9456	0.8249	0.8124	0.8178	0.945
No log	6.0	120	0.2083	0.9432	0.8113	0.8173	0.8136	0.9429
No log	6.5	130	0.2105	0.9441	0.8355	0.8132	0.8236	0.9436
No log	7.0	140	0.2083	0.9439	0.8312	0.8207	0.8253	0.9439
No log	7.5	150	0.2145	0.9447	0.8293	0.8051	0.8161	0.9437

Framework versions

Transformers 4.35.2
Pytorch 2.0.1+cu117
Datasets 2.18.0
Tokenizers 0.15.0