Tommert25/multibert2209_scrambled

multibertfinetuned2209

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

• Loss: 0.3973
• Precision: 0.7567
• Recall: 0.7607
• F1: 0.7587
• Accuracy: 0.9064

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: 5e-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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	118	0.4058	0.7597	0.7343	0.7468	0.9032
No log	2.0	236	0.3973	0.7567	0.7607	0.7587	0.9064
No log	3.0	354	0.4153	0.7540	0.7677	0.7608	0.9062
No log	4.0	472	0.4656	0.7645	0.7466	0.7555	0.9082
0.0692	5.0	590	0.4940	0.7594	0.7554	0.7574	0.9043
0.0692	6.0	708	0.5446	0.7668	0.7484	0.7575	0.9059
0.0692	7.0	826	0.5732	0.7818	0.7420	0.7613	0.9069
0.0692	8.0	944	0.5668	0.7844	0.7431	0.7632	0.9082

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3