Hugging Face's logo

mateussj
/
oracle_class_vert_v3

Text Classification
Transformers
Safetensors
bert
Generated from Trainer
Inference Endpoints
Model card Files Community
oracle_class_vert_v3 / README.md
mateussj's picture
mateussj
oracle-corejur/bert_oracle_class_final_dataset_vert_v2
e1d8260 verified
preview code
|
raw
history blame contribute delete
No virus
1.94 kB
metadata
license: mit
base_model: neuralmind/bert-base-portuguese-cased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - accuracy
  - f1
model-index:
  - name: oracle_class_vert_v3
    results: []

Visualize in Weights & Biases

oracle_class_vert_v3

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: 0.5665
  • Precision: 0.8326
  • Recall: 0.8336
  • Accuracy: 0.8477
  • F1: 0.8310

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: 6e-05
  • train_batch_size: 24
  • eval_batch_size: 24
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Precision Recall Accuracy F1
0.5484 1.4589 550 0.5911 0.8063 0.8028 0.8308 0.8010
0.3253 2.9178 1100 0.5665 0.8326 0.8336 0.8477 0.8310

Framework versions

  • Transformers 4.43.1
  • Pytorch 2.2.0
  • Datasets 2.20.0
  • Tokenizers 0.19.1