xlm-roberta-base-finetuned-panx-hi-ta

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

• Loss: 0.4635
• F1: 0.8674

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: 24
• eval_batch_size: 24
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
0.4856	1.0	372	0.2799	0.7713
0.2483	2.0	744	0.2545	0.8067
0.1844	3.0	1116	0.2355	0.8147
0.1486	4.0	1488	0.2546	0.8136
0.1373	5.0	1860	0.2560	0.8167
0.0839	6.0	2232	0.2699	0.8232
0.0658	7.0	2604	0.3030	0.8209
0.05	8.0	2976	0.3625	0.8458
0.0382	9.0	3348	0.3777	0.8445
0.031	10.0	3720	0.3732	0.8514
0.0246	11.0	4092	0.3994	0.8579
0.0144	12.0	4464	0.4308	0.8507
0.0137	13.0	4836	0.4203	0.8499
0.0085	14.0	5208	0.4428	0.8613
0.0067	15.0	5580	0.4589	0.8563
0.0062	16.0	5952	0.4375	0.8609
0.0047	17.0	6324	0.4448	0.8610
0.0026	18.0	6696	0.4540	0.8624
0.0017	19.0	7068	0.4645	0.8658
0.0021	20.0	7440	0.4635	0.8674

Framework versions

• Transformers 4.33.1
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3