16_combo_webscrap_1709_v2_reduce_others

This model is a fine-tuned version of bert-base-multilingual-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.1501
• Accuracy: 0.9636

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: 1e-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: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	363	1.0481	0.7263
1.5287	2.0	726	0.5613	0.8655
0.6856	3.0	1089	0.3666	0.9121
0.6856	4.0	1452	0.2880	0.9284
0.4313	5.0	1815	0.2187	0.9464
0.3097	6.0	2178	0.1992	0.9505
0.2454	7.0	2541	0.1627	0.9598
0.2454	8.0	2904	0.1501	0.9636

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu117
• Datasets 2.14.5
• Tokenizers 0.13.3