adriansanz/fm-tc-end_mix_xml_labelsrt

fm-tc-end_mix_xml

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: 0.2061
• Accuracy: 0.97
• Precision: 0.9706
• Recall: 0.9700
• F1: 0.9700

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
0.6642	1.0	1188	0.4540	0.894	0.8983	0.8940	0.8938
0.4074	2.0	2376	0.4206	0.918	0.9241	0.9180	0.9183
0.2403	3.0	3564	0.4380	0.918	0.9221	0.9180	0.9165
0.1625	4.0	4752	0.4773	0.926	0.9297	0.9260	0.9262
0.0991	5.0	5940	0.2999	0.952	0.9536	0.9520	0.9521
0.0549	6.0	7128	0.2217	0.966	0.9671	0.966	0.9659
0.0226	7.0	8316	0.2770	0.964	0.9650	0.9640	0.9637
0.0154	8.0	9504	0.2061	0.97	0.9706	0.9700	0.9700
0.0061	9.0	10692	0.2372	0.97	0.9706	0.9700	0.9699
0.0048	10.0	11880	0.2419	0.964	0.9650	0.9640	0.9640

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.19.2
• Tokenizers 0.19.1