bangla-hatespeech-analysis-banglaBert-Base

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

• Loss: 0.4189
• Accuracy: 0.8226

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.466	0.2527	1000	0.4682	0.7893
0.4273	0.5053	2000	0.4433	0.8059
0.4426	0.7580	3000	0.4244	0.8104
0.4141	1.0106	4000	0.4794	0.7972
0.4109	1.2633	5000	0.4585	0.8112
0.4181	1.5159	6000	0.4249	0.8084
0.3519	1.7686	7000	0.4134	0.8186
0.3572	2.0212	8000	0.4118	0.8217
0.3251	2.2739	9000	0.4136	0.8184
0.3145	2.5265	10000	0.4335	0.8190
0.3247	2.7792	11000	0.4189	0.8226

Framework versions

• Transformers 4.50.0
• Pytorch 2.6.0+cu124
• Tokenizers 0.21.1