bert-base-multilingual-uncased-sentiment-finetuned-mnli

This model is a fine-tuned version of nlptown/bert-base-multilingual-uncased-sentiment on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.5330
• Accuracy: 0.7902

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: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5568	1.0	1080	0.5330	0.7902
0.4713	2.0	2160	0.5633	0.7875
0.3791	3.0	3240	0.6680	0.7824
0.2967	4.0	4320	0.8067	0.7624
0.2121	5.0	5400	0.9723	0.7624
0.1511	6.0	6480	1.1602	0.7629
0.1277	7.0	7560	1.4037	0.7736
0.0931	8.0	8640	1.5388	0.7675
0.0768	9.0	9720	2.0003	0.7330
0.0457	10.0	10800	1.8301	0.7756
0.0383	11.0	11880	1.9697	0.7701
0.0286	12.0	12960	2.0533	0.7756
0.0175	13.0	14040	2.2299	0.7594
0.0101	14.0	15120	2.1549	0.7749
0.0055	15.0	16200	2.2199	0.7703

Framework versions

• Transformers 4.41.0
• Pytorch 2.3.0+cu121
• Datasets 2.19.1
• Tokenizers 0.19.1