cvalenciaunivalle/sarcasmo-beto

Sarcasmo BETO

Fine-tuning de dccuchile/bert-base-spanish-wwm-cased sobre el dataset Sarcastic_spanish_dataset (Ernesto-1997). Tarea: clasificación binaria (sarcasmo / no_sarcasmo). Taller PLN - Maestría Ingeniería de Sistemas, Univalle.

Métricas en test

Batch	F1	Precision	Recall	Accuracy
bs8	0.8920	0.9116	0.8732	0.9170
bs16	0.8935	0.9055	0.8819	0.9175
bs32 ⭐	0.8954	0.9002	0.8906	0.9183

Uso

from transformers import AutoModelForSequenceClassification, AutoTokenizer, pipeline

m = AutoModelForSequenceClassification.from_pretrained("cvalenciaunivalle/sarcasmo-beto", subfolder="bs32")
t = AutoTokenizer.from_pretrained("cvalenciaunivalle/sarcasmo-beto", subfolder="bs32")
pipe = pipeline("text-classification", model=m, tokenizer=t)
pipe("¡Qué divertido, otro lunes de trabajo!")

Detalles

• Modelo base: dccuchile/bert-base-spanish-wwm-cased
• Dataset: Ernesto-1997/Sarcastic_spanish_dataset
• Trainer + EarlyStopping
• Hardware: Kaggle GPU
• Compañeros: Laura Valero, Cristian Valencia