10k_test3_nli_finetuned_bert

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: 0.9076
• Accuracy: 0.6693
• F1 Weighted: 0.6690
• Precision Weighted: 0.6889
• Recall Weighted: 0.6693

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: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 8
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1 Weighted	Precision Weighted	Recall Weighted
0.8753	1.0	312	0.9773	0.5313	0.5138	0.6276	0.5313
0.6746	2.0	625	0.7384	0.692	0.6927	0.6954	0.692
0.4239	3.0	936	0.9076	0.6693	0.6690	0.6889	0.6693

Framework versions

• Transformers 4.27.2
• Pytorch 2.0.0+cu117
• Datasets 2.10.1
• Tokenizers 0.13.2