metadata

license: mit
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: tmvar_0.0001
    results: []

tmvar_0.0001

This model is a fine-tuned version of microsoft/BiomedNLP-PubMedBERT-base-uncased-abstract-fulltext on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.0162
Precision: 0.8877
Recall: 0.8973
F1: 0.8925
Accuracy: 0.9971

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: 0.0001
train_batch_size: 16
eval_batch_size: 16
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
training_steps: 500

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.2263	1.47	25	0.0776	0.0	0.0	0.0	0.9843
0.05	2.94	50	0.0400	0.2868	0.4216	0.3414	0.9872
0.0271	4.41	75	0.0219	0.5381	0.6486	0.5882	0.9925
0.0108	5.88	100	0.0132	0.8324	0.8324	0.8324	0.9965
0.0029	7.35	125	0.0107	0.8934	0.9514	0.9215	0.9979
0.0025	8.82	150	0.0123	0.8691	0.8973	0.8830	0.9972
0.0011	10.29	175	0.0127	0.8579	0.9135	0.8848	0.9969
0.0006	11.76	200	0.0102	0.8969	0.9405	0.9182	0.9981
0.0005	13.24	225	0.0118	0.8942	0.9135	0.9037	0.9978
0.0005	14.71	250	0.0106	0.8768	0.9622	0.9175	0.9981
0.0015	16.18	275	0.0119	0.855	0.9243	0.8883	0.9976
0.0006	17.65	300	0.0134	0.8814	0.9243	0.9024	0.9977
0.0004	19.12	325	0.0177	0.8617	0.8757	0.8686	0.9969
0.0003	20.59	350	0.0162	0.8877	0.8973	0.8925	0.9971

Framework versions

Transformers 4.27.4
Pytorch 2.0.0+cu118
Datasets 2.11.0
Tokenizers 0.13.2