langid-ner-xlm-v-base

This model is a fine-tuned version of facebook/xlm-v-base on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.4352
• Precision: 0.7966
• Recall: 0.7905
• F1: 0.7935
• Accuracy: 0.8978

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: 16
• seed: 42
• 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: linear
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	72	1.5446	0.5671	0.5671	0.5671	0.5366
No log	2.0	144	1.0001	0.6146	0.6367	0.6255	0.7876
No log	3.0	216	0.7811	0.6488	0.6493	0.6491	0.8160
No log	4.0	288	0.6540	0.7165	0.7315	0.7240	0.8648
No log	5.0	360	0.5871	0.7323	0.7366	0.7344	0.8725
No log	6.0	432	0.5591	0.7363	0.7424	0.7393	0.8719
0.966	7.0	504	0.5282	0.7454	0.7466	0.7460	0.8812
0.966	8.0	576	0.5095	0.7438	0.7525	0.7481	0.8792
0.966	9.0	648	0.4940	0.7444	0.7525	0.7484	0.8792
0.966	10.0	720	0.4923	0.7432	0.7525	0.7478	0.8802

Framework versions

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