metadata
license: mit
tags:
- generated_from_trainer
model-index:
- name: bert_base_tcm_0.6
results: []
bert_base_tcm_0.6
This model is a fine-tuned version of neuralmind/bert-base-portuguese-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0109
- Criterio Julgamento Precision: 0.8409
- Criterio Julgamento Recall: 0.925
- Criterio Julgamento F1: 0.8810
- Criterio Julgamento Number: 80
- Data Sessao Precision: 0.7838
- Data Sessao Recall: 0.8056
- Data Sessao F1: 0.7945
- Data Sessao Number: 36
- Modalidade Licitacao Precision: 0.9517
- Modalidade Licitacao Recall: 0.9718
- Modalidade Licitacao F1: 0.9617
- Modalidade Licitacao Number: 284
- Numero Exercicio Precision: 0.9706
- Numero Exercicio Recall: 0.9925
- Numero Exercicio F1: 0.9814
- Numero Exercicio Number: 133
- Objeto Licitacao Precision: 0.6143
- Objeto Licitacao Recall: 0.7544
- Objeto Licitacao F1: 0.6772
- Objeto Licitacao Number: 57
- Valor Objeto Precision: 0.8571
- Valor Objeto Recall: 1.0
- Valor Objeto F1: 0.9231
- Valor Objeto Number: 6
- Overall Precision: 0.8917
- Overall Recall: 0.9396
- Overall F1: 0.9150
- Overall Accuracy: 0.9980
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: 5e-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: 5.0
Training results
Framework versions
- Transformers 4.20.0.dev0
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1