tes1-UASNLP2

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

• Loss: 0.3413
• Accuracy: 0.8705
• Precision: 0.8858
• Recall: 0.8803
• F1: 0.8830

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: 64
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.4596	1.2121	100	0.3631	0.8554	0.8767	0.8605	0.8685
0.254	2.4242	200	0.3413	0.8705	0.8858	0.8803	0.8830
0.1674	3.6364	300	0.3847	0.8793	0.8758	0.9118	0.8934
0.0968	4.8485	400	0.4927	0.8759	0.9145	0.8564	0.8845
0.0614	6.0606	500	0.5308	0.8721	0.8748	0.8981	0.8863
0.0418	7.2727	600	0.6098	0.8759	0.8988	0.8748	0.8867
0.0296	8.4848	700	0.6507	0.8751	0.8910	0.8830	0.8870
0.0183	9.6970	800	0.6822	0.8789	0.8944	0.8865	0.8904

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Tokenizers 0.19.1