bengali-sentiment-v2

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

• Loss: 0.3462
• F1 Macro: 0.7174
• F1 Micro: 0.7611
• Precision Macro: 0.8027
• Recall Macro: 0.6725
• Exact Match Accuracy: 0.5171

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: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	F1 Macro	F1 Micro	Precision Macro	Recall Macro	Exact Match Accuracy
0.3892	1.0	204	0.3781	0.5848	0.6937	0.8163	0.5252	0.3943
0.3557	2.0	408	0.3613	0.6813	0.7429	0.8102	0.6203	0.4829
0.3004	3.0	612	0.3462	0.7174	0.7611	0.8027	0.6725	0.5171

Framework versions

• Transformers 5.21.0
• Pytorch 2.15.0+cu132
• Datasets 5.1.0
• Tokenizers 0.23.3