metadata

base_model: MMG/mlm-spanish-roberta-base
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: roberta-finetuned-gesture-prediction-es
    results: []

roberta-finetuned-gesture-prediction-es

This model is a fine-tuned version of MMG/mlm-spanish-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 1.7706
Accuracy: 0.7223
Precision: 0.7215
Recall: 0.7223
F1: 0.7156

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: 2e-05
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
num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
1.8523	1.0	102	1.2237	0.6618	0.6205	0.6618	0.6316
1.0093	2.0	204	1.1357	0.6886	0.6715	0.6886	0.6663
0.6999	3.0	306	1.1758	0.6884	0.7008	0.6884	0.6763
0.4872	4.0	408	1.1398	0.6955	0.6982	0.6955	0.6839
0.3198	5.0	510	1.2017	0.7096	0.7112	0.7096	0.7059
0.2414	6.0	612	1.2819	0.7152	0.7101	0.7152	0.7049
0.1676	7.0	714	1.3279	0.7299	0.7272	0.7299	0.7221
0.1245	8.0	816	1.4593	0.7098	0.7078	0.7098	0.7011
0.0843	9.0	918	1.5682	0.7134	0.7131	0.7134	0.7063
0.0636	10.0	1020	1.5447	0.7195	0.7161	0.7195	0.7128
0.0464	11.0	1122	1.6686	0.7118	0.7164	0.7118	0.7050
0.0367	12.0	1224	1.6438	0.7251	0.7252	0.7251	0.7181
0.0292	13.0	1326	1.6803	0.7232	0.7199	0.7232	0.7170
0.0227	14.0	1428	1.6852	0.7217	0.7193	0.7217	0.7157
0.0155	15.0	1530	1.7753	0.7219	0.7245	0.7219	0.7156
0.0123	16.0	1632	1.7875	0.7157	0.7149	0.7157	0.7085
0.0102	17.0	1734	1.7649	0.7159	0.7148	0.7159	0.7095
0.0076	18.0	1836	1.7740	0.7204	0.7201	0.7204	0.7141
0.0074	19.0	1938	1.7674	0.7244	0.7235	0.7244	0.7177
0.0061	20.0	2040	1.7706	0.7223	0.7215	0.7223	0.7156

Framework versions

Transformers 4.35.2
Pytorch 2.1.0+cu121
Datasets 2.16.1
Tokenizers 0.15.0