HODravidianLangTech

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: 0.5859
• F1: 0.6908

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: 1e-06
• train_batch_size: 16
• eval_batch_size: 16
• seed: 1234
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	F1
No log	1.0	100	0.6926	0.3377
No log	2.0	200	0.6916	0.5490
No log	3.0	300	0.6856	0.6050
No log	4.0	400	0.6701	0.6287
0.6833	5.0	500	0.6601	0.6396
0.6833	6.0	600	0.6511	0.6466
0.6833	7.0	700	0.6447	0.6458
0.6833	8.0	800	0.6250	0.6560
0.6833	9.0	900	0.6113	0.6516
0.624	10.0	1000	0.6051	0.6658
0.624	11.0	1100	0.6075	0.6567
0.624	12.0	1200	0.6038	0.6671
0.624	13.0	1300	0.5997	0.6716
0.624	14.0	1400	0.5949	0.6805
0.5739	15.0	1500	0.5958	0.6885
0.5739	16.0	1600	0.5924	0.6905
0.5739	17.0	1700	0.5957	0.6875
0.5739	18.0	1800	0.5839	0.6976
0.5739	19.0	1900	0.5865	0.6908
0.5598	20.0	2000	0.5859	0.6908

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu118
• Datasets 2.15.0
• Tokenizers 0.15.0