adriansanz/150524_15ep

080524_epoch_13

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: 0.8371
• Accuracy: 0.8151
• Precision: 0.8509
• Recall: 0.8151
• F1: 0.8103
• Ratio: 0.6597

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: 10
• eval_batch_size: 2
• seed: 47
• gradient_accumulation_steps: 2
• total_train_batch_size: 20
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.06
• lr_scheduler_warmup_steps: 4
• num_epochs: 1
• label_smoothing_factor: 0.1

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Ratio
0.3002	0.1176	10	0.8662	0.8151	0.8509	0.8151	0.8103	0.6597
0.3026	0.2353	20	0.7930	0.8277	0.8516	0.8277	0.8248	0.6303
0.2933	0.3529	30	0.7946	0.8277	0.8484	0.8277	0.8251	0.6218
0.2921	0.4706	40	0.8687	0.8151	0.8509	0.8151	0.8103	0.6597
0.2947	0.5882	50	0.8540	0.8109	0.8442	0.8109	0.8062	0.6555
0.3148	0.7059	60	0.8454	0.8151	0.8469	0.8151	0.8108	0.6513
0.3221	0.8235	70	0.8642	0.8151	0.8509	0.8151	0.8103	0.6597
0.316	0.9412	80	0.8389	0.8151	0.8509	0.8151	0.8103	0.6597

Framework versions

• Transformers 4.40.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1