finetuned_roberta-base-uncased

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: 1.4799
• Accuracy: 0.6519

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: 32
• eval_batch_size: 32
• seed: 42
• 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	Accuracy
1.372	1.0	102	1.3643	0.3375
1.1591	2.0	204	1.1988	0.4830
0.9623	3.0	306	1.0802	0.5694
0.7766	4.0	408	0.9885	0.6237
0.7336	5.0	510	1.0393	0.6120
0.6284	6.0	612	1.1150	0.6392
0.3616	7.0	714	1.2183	0.6402
0.3526	8.0	816	1.2362	0.6305
0.3151	9.0	918	1.3058	0.6372
0.3035	10.0	1020	1.2966	0.6343
0.2458	11.0	1122	1.3752	0.6508
0.2469	12.0	1224	1.4557	0.6557
0.2039	13.0	1326	1.5541	0.6372
0.1691	14.0	1428	1.5308	0.6343
0.1455	15.0	1530	1.6339	0.6421
0.1716	16.0	1632	1.6843	0.6392
0.1698	17.0	1734	1.6802	0.6479
0.2009	18.0	1836	1.6544	0.6479
0.1415	19.0	1938	1.6759	0.6518
0.1616	20.0	2040	1.6833	0.6508

Framework versions

• Transformers 4.26.1
• Pytorch 1.13.1+cu116
• Datasets 2.10.1
• Tokenizers 0.13.2