20231008-1-distilbert-base-multilingual-cased-new

This model is a fine-tuned version of distilbert-base-multilingual-cased on an unknown dataset. It achieves the following results on the evaluation set:

• Accuracy: 0.5583
• Loss: 1.8644

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
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss
3.0435	1.82	200	0.4356	2.5262
2.5692	3.64	400	0.4205	2.6711
2.3352	5.45	600	0.4841	2.1151
2.2573	7.27	800	0.5188	2.1011
2.1775	9.09	1000	0.4776	2.1433
2.0626	10.91	1200	0.5370	2.0399
2.0071	12.73	1400	0.5732	1.8666
1.9289	14.55	1600	0.6042	1.6883
1.9511	16.36	1800	0.6228	1.7540
1.8715	18.18	2000	0.6299	1.6179
1.839	20.0	2200	0.5583	1.8644

Framework versions

• Transformers 4.34.0
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.14.1