arq-xlm

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: 0.6149
• F1-micro: 0.0
• F1-macro: 0.0

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: 1.5e-06
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 18
• num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss	F1-micro	F1-macro
0.7029	1.0	45	0.6908	0.2746	0.1565
0.6612	2.0	90	0.6355	0.0177	0.0184
0.6397	3.0	135	0.6188	0.0062	0.0050
0.634	4.0	180	0.6149	0.0	0.0

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1