license: apache-2.0
base_model: distilbert-base-multilingual-cased
tags:
- generated_from_trainer
- pop science
model-index:
- name: results
results: []
language:
- es
metrics:
- f1
- roc_auc
pipeline_tag: text-classification
datasets:
- teoremaclon/popscitweetsbyarea
library_name: transformers
results
This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1582
- Roc Auc: 0.7868
- Hamming Loss: 0.0625
- F1 Score: 0.6231
Model description
Label interpretation: 'astronomía y espacio' (astronomy and space): 0, 'matemáticas' (mathematics): 1, 'física' (physics): 2, 'biología' (biology): 3, 'medicina y salud' (health and medicine): 4, 'tecnología' (technology): 5, 'química' (chemistry): 6, 'historia de la ciencia' (history of science): 7, 'ingeniería' (engineering): 8, 'computación' (computation): 9, 'ciencias de la tierra' (earth science): 10, 'materia y energia' (matter and energy): 11, 'psicología' (psychology): 12, 'invitación a evento o a recursos' (invitation to event or resources): 13, 'efeméride' (anniversary/day of): 14, 'mujeres en la ciencia' (women in science): 15, 'cultura pop' (pop culture): 16, 'otro' (other): 17
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: 3
Training results
Framework versions
- Transformers 4.39.1
- Pytorch 2.2.1+cu121
- Tokenizers 0.15.2