metadata

library_name: transformers
license: mit
base_model: belisards/congretimbau
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
  - recall
  - precision
model-index:
  - name: belisards/congretimbau
    results: []

belisards/congretimbau

This model is a fine-tuned version of belisards/congretimbau on the None dataset. It achieves the following results on the evaluation set:

Loss: 0.1176
Accuracy: 0.8027
F1: 0.7358
Recall: 0.7544
Precision: 0.7236

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: 1e-05
train_batch_size: 64
eval_batch_size: 64
seed: 5151
optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 200
num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Recall	Precision
0.3405	1.0	18	0.2083	0.7232	0.4751	0.5104	0.5393
0.1467	2.0	36	0.1258	0.4107	0.4105	0.5463	0.5486
0.1198	3.0	54	0.1127	0.6607	0.5988	0.6140	0.5964
0.107	4.0	72	0.0999	0.6696	0.6339	0.6762	0.6380
0.0987	5.0	90	0.0943	0.6339	0.6113	0.6745	0.6339
0.0911	6.0	108	0.0930	0.6875	0.6492	0.6882	0.6492
0.078	7.0	126	0.0953	0.7321	0.6883	0.7183	0.6805
0.0671	8.0	144	0.0934	0.7232	0.6850	0.7235	0.6798
0.0534	9.0	162	0.1065	0.8036	0.7441	0.7441	0.7441
0.0355	10.0	180	0.1363	0.8214	0.7724	0.7786	0.7670
0.0263	11.0	198	0.1411	0.8214	0.7724	0.7786	0.7670
0.013	12.0	216	0.2712	0.8214	0.7560	0.7449	0.7710
0.0074	13.0	234	0.3198	0.7946	0.7294	0.7268	0.7321

Framework versions

Transformers 4.47.0
Pytorch 2.5.1+cu121
Datasets 3.2.0
Tokenizers 0.21.0