adriansanz/2504separado1

2504v4-8ep

This model is a fine-tuned version of projecte-aina/roberta-base-ca-v2-cased-te on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5711
• Accuracy: 0.8487
• Precision: 0.8523
• Recall: 0.8487
• F1: 0.8484
• Ratio: 0.5504

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: 2e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06
• num_epochs: 4
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Ratio
2.0709	0.9870	38	0.8049	0.7059	0.7073	0.7059	0.7054	0.4580
0.7325	2.0	77	0.6190	0.8067	0.8081	0.8067	0.8065	0.5336
0.6249	2.9870	115	0.5998	0.8109	0.8230	0.8109	0.8091	0.5966
0.5768	3.9481	152	0.5711	0.8487	0.8523	0.8487	0.8484	0.5504

Framework versions

• Transformers 4.40.0
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.19.1