metadata

language:
  - es
metrics:
  - f1

Model Description

This model is a fine-tuned version of PlanTL-GOB-ES/roberta-base-bne to detect suicidal ideation/behavior on public comments (forums, twitter, etc.) in Spanish language.

How to use

from transformers import pipeline


model_name= 'dariolopez/roberta-base-bne-finetuned-suicide-es'
pipe = pipeline("text-classification", model=model_name, device=0)

pipe("Quiero acabar con todo. No merece la pena vivir.")
[{'label': 'Suicide', 'score': 0.9999703168869019}]

pipe("El partido de fútbol fue igualado, disfrutamos mucho jugando juntos.")
[{'label': 'Non-Suicide', 'score': 0.999990701675415}]