Instructions to use csala23/JustinIA with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use csala23/JustinIA with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="csala23/JustinIA")# Load model directly from transformers import AutoTokenizer, AutoModelForSequenceClassification tokenizer = AutoTokenizer.from_pretrained("csala23/JustinIA") model = AutoModelForSequenceClassification.from_pretrained("csala23/JustinIA") - Notebooks
- Google Colab
- Kaggle
Model Card for Model ID
JustinIA V1.0 2023
Model Details
Modelo Base: roBERTa Dataset: Propietario Publicado: 31 de marzo del 2023
Model Description
Modelo de Inteligencia Artificial para la tarea de clasificación de textos juridicos en español. Clasificación de acuerdo al tema de una sentencia en materia de justicia administrativa
- Developed by: Salazar Flores Carlos Francisco
- Shared by : Salazar Flores Carlos Francisco
- Model type: Transformers
- Language(s) (NLP): Español Méxicano
- License:
- Finetuned from model : roberta-base-bne
Model Sources
- Repository: BSC-TeMU/roberta-base-bne
Training Details
5 ciclos de entrenamiento, con una precisión máxima de 28.52%
-Epoch Training Loss Validation Loss Accuracy -1 1.982000 2.498654 0.274072 -2 1.730200 2.486177 0.278226 -3 1.551800 2.573319 0.273923 -4 1.359700 2.571429 0.285199 -5 1.208200 2.623226 0.283863
-TrainOutput(global_step=9930, training_loss=1.6167555717783273, metrics={'train_runtime': 1732.3659, 'train_samples_per_second': 57.32, 'train_steps_per_second': 5.732, 'total_flos': 2877096870824520.0, 'train_loss': 1.6167555717783273, 'epoch': 5.0})
Metrics
Precisión
Results
Modelo de Inteligencia Articial basado en arquitectura de transformadores, entrenado para la tarea de clasificación de textos en español en materia de Justicia Administrativa.
- Downloads last month
- 4