robeczech_lr3e-05_bs16_train287

This model is a fine-tuned version of FacebookAI/xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1179
• Precision: 0.9454
• Recall: 0.9595
• F1: 0.9524
• Accuracy: 0.9714

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 30

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	18	1.1550	1.0	0.0005	0.0010	0.5668
No log	2.0	36	0.4725	0.7099	0.7006	0.7052	0.8587
No log	3.0	54	0.2293	0.8740	0.8643	0.8691	0.9351
No log	4.0	72	0.1474	0.9224	0.9126	0.9175	0.9565
No log	5.0	90	0.1210	0.9457	0.9411	0.9434	0.9697
No log	6.0	108	0.1212	0.9409	0.9382	0.9396	0.9674
No log	7.0	126	0.1067	0.9540	0.9517	0.9529	0.9740
No log	8.0	144	0.0918	0.9574	0.9551	0.9562	0.9753
No log	9.0	162	0.1076	0.9549	0.9517	0.9533	0.9749
No log	10.0	180	0.0990	0.9599	0.9585	0.9592	0.9774
No log	11.0	198	0.1027	0.9673	0.9570	0.9621	0.9778

Framework versions

• Transformers 4.51.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1