metadata
license: mit
base_model: FacebookAI/xlm-roberta-base
tags:
- generated_from_trainer
metrics:
- accuracy
- precision
- recall
- f1
model-index:
- name: soluciona_fm_tcv_1
results: []
mapping
'Accessibilitat': 0,
'Arbrat': 1,
'Desguassos i embornals': 2,
'Enllumenat públic': 3,
'Escombraries acumulades': 4,
'Fuites d\'aigua': 5,
'Grafits': 6,
'Mobiliari urbà': 7,
'Obres inacabades': 8,
'Parcs i jardins': 9,
'Plagues': 10,
'Senyalització vial': 11,
'Sorolls': 12,
'Sots a la calçada': 13,
'Vehicles abandonats': 14,
'Voreres danyades': 15
soluciona_fm_tcv_1
This model is a fine-tuned version of FacebookAI/xlm-roberta-base on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0260
- Accuracy: 0.996
- Precision: 0.9962
- Recall: 0.9960
- F1: 0.9960
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: 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: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 |
---|---|---|---|---|---|---|---|
0.9405 | 1.0 | 675 | 0.1074 | 0.978 | 0.9793 | 0.9778 | 0.9780 |
0.106 | 2.0 | 1350 | 0.0876 | 0.986 | 0.9866 | 0.9861 | 0.9860 |
0.0237 | 3.0 | 2025 | 0.0491 | 0.992 | 0.9923 | 0.9920 | 0.9920 |
0.0181 | 4.0 | 2700 | 0.0620 | 0.988 | 0.9897 | 0.9879 | 0.9880 |
0.0206 | 5.0 | 3375 | 0.0332 | 0.994 | 0.9943 | 0.9940 | 0.9940 |
0.0071 | 6.0 | 4050 | 0.0260 | 0.996 | 0.9962 | 0.9960 | 0.9960 |
0.0012 | 7.0 | 4725 | 0.0416 | 0.996 | 0.9962 | 0.9960 | 0.9960 |
0.009 | 8.0 | 5400 | 0.0723 | 0.994 | 0.9943 | 0.9940 | 0.9940 |
0.0001 | 9.0 | 6075 | 0.0447 | 0.996 | 0.9962 | 0.9960 | 0.9960 |
0.0001 | 10.0 | 6750 | 0.0469 | 0.996 | 0.9962 | 0.9960 | 0.9960 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.0+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1