xlm-roberta-base-finetuned-pos-tr

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

• Loss: 0.2340
• Precision: 0.9437
• Recall: 0.9445
• F1: 0.9441
• Accuracy: 0.9654

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: 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	Precision	Recall	F1	Accuracy
0.3157	1.0	836	0.1525	0.9241	0.9352	0.9296	0.9552
0.1291	2.0	1672	0.1430	0.9286	0.9337	0.9311	0.9573
0.1021	3.0	2508	0.1231	0.9344	0.9405	0.9374	0.9617
0.0868	4.0	3344	0.1220	0.9393	0.9422	0.9407	0.9635
0.0723	5.0	4180	0.1224	0.9419	0.9457	0.9438	0.9657
0.0634	6.0	5016	0.1329	0.9413	0.9431	0.9422	0.9641
0.0545	7.0	5852	0.1441	0.9422	0.9440	0.9431	0.9646
0.0476	8.0	6688	0.1541	0.9422	0.9431	0.9427	0.9643
0.0419	9.0	7524	0.1603	0.9428	0.9455	0.9442	0.9652
0.0357	10.0	8360	0.1693	0.9439	0.9455	0.9447	0.9649
0.0313	11.0	9196	0.1770	0.9441	0.9451	0.9446	0.9653
0.0274	12.0	10032	0.1880	0.9414	0.9435	0.9425	0.9645
0.0235	13.0	10868	0.1986	0.9452	0.9441	0.9446	0.9653
0.021	14.0	11704	0.2081	0.9436	0.9439	0.9438	0.9650
0.0177	15.0	12540	0.2116	0.9433	0.9440	0.9437	0.9651
0.0153	16.0	13376	0.2174	0.9457	0.9452	0.9454	0.9662
0.0132	17.0	14212	0.2252	0.9431	0.9441	0.9436	0.9649
0.0126	18.0	15048	0.2292	0.9432	0.9441	0.9437	0.9653
0.0106	19.0	15884	0.2338	0.9440	0.9442	0.9441	0.9654
0.0106	20.0	16720	0.2340	0.9437	0.9445	0.9441	0.9654

Framework versions

• Transformers 4.34.1
• Pytorch 2.2.1+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1