bengali_pos_v1_200000

This model is a fine-tuned version of mHossain/bengali_pos_v1 on the pos_tag_100k dataset. It achieves the following results on the evaluation set:

• Loss: 0.5802
• Precision: 0.7728
• Recall: 0.7752
• F1: 0.7740
• Accuracy: 0.8326

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.6183	1.0	22500	0.6032	0.7546	0.7570	0.7558	0.8193
0.5138	2.0	45000	0.5763	0.7691	0.7694	0.7692	0.8292
0.4448	3.0	67500	0.5802	0.7728	0.7752	0.7740	0.8326

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0