young_old_classificator

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

• Loss: 0.2153
• Accuracy: 0.9647
• F1: 0.9646
• Precision: 0.9670
• Recall: 0.9647

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: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20
• mixed_precision_training: Native AMP
• label_smoothing_factor: 0.15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.7046	1.0	21	0.6945	0.5175	0.3530	0.2679	0.5175
0.6918	2.0	42	0.6841	0.5175	0.3530	0.2679	0.5175
0.6247	3.0	63	0.5395	0.7018	0.6799	0.7978	0.7018
0.405	4.0	84	0.3227	0.8947	0.8945	0.9031	0.8947
0.2622	5.0	105	0.5418	0.8596	0.8586	0.8772	0.8596
0.185	6.0	126	0.3322	0.9211	0.9210	0.9245	0.9211
0.1968	7.0	147	0.2518	0.9386	0.9386	0.9399	0.9386
0.2289	8.0	168	0.2705	0.9386	0.9386	0.9399	0.9386
0.1514	9.0	189	0.1881	0.9561	0.9562	0.9575	0.9561
0.072	10.0	210	0.1173	0.9649	0.9649	0.9655	0.9649
0.066	11.0	231	0.1214	0.9737	0.9737	0.9750	0.9737
0.0299	12.0	252	0.1070	0.9737	0.9737	0.9738	0.9737
0.0651	13.0	273	0.1556	0.9649	0.9649	0.9655	0.9649
0.0889	14.0	294	0.1653	0.9737	0.9737	0.9750	0.9737

Framework versions

• Transformers 4.57.2
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1