metadata

base_model: dccuchile/bert-base-spanish-wwm-cased
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: FNST_trad_2h
    results: []

FNST_trad_2h

This model is a fine-tuned version of dccuchile/bert-base-spanish-wwm-cased on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.3483
Accuracy: 0.6981
F1: 0.6890

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: 16
eval_batch_size: 16
seed: 42
gradient_accumulation_steps: 2
total_train_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 8
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
0.9472	0.32	500	0.8488	0.6418	0.6296
0.8362	0.64	1000	0.7882	0.6641	0.6492
0.7964	0.96	1500	0.7619	0.6821	0.6745
0.6268	1.28	2000	0.7757	0.6909	0.6829
0.6134	1.6	2500	0.7786	0.6857	0.6757
0.6236	1.92	3000	0.7758	0.6830	0.6713
0.456	2.24	3500	0.9266	0.6779	0.6688
0.4198	2.56	4000	0.8362	0.6859	0.6796
0.4334	2.88	4500	0.8705	0.6916	0.6813
0.3266	3.2	5000	1.0376	0.6934	0.6841
0.2825	3.52	5500	1.0690	0.6940	0.6821
0.294	3.84	6000	1.1104	0.6896	0.6812
0.2493	4.16	6500	1.3248	0.6848	0.6764
0.2085	4.48	7000	1.1945	0.6869	0.6789
0.2202	4.8	7500	1.2104	0.6853	0.6739
0.2011	5.12	8000	1.4272	0.6932	0.6853
0.1724	5.44	8500	1.3933	0.6823	0.6682
0.1873	5.76	9000	1.3483	0.6981	0.6890
0.1791	6.08	9500	1.5552	0.6815	0.6704
0.1527	6.4	10000	1.4202	0.6873	0.6754
0.1631	6.72	10500	1.6333	0.6727	0.6610
0.1635	7.04	11000	1.6169	0.6850	0.6696
0.1341	7.36	11500	1.5840	0.6871	0.6785
0.1513	7.68	12000	1.4788	0.6848	0.6739

Framework versions

Transformers 4.37.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.1