pritamdeka/mDeBERTa-v3-base-Assamese

temp_assamese

This model is a fine-tuned version of microsoft/mdeberta-v3-base on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.8132
• Accuracy: 0.8287

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: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 1.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.4466	0.0931	5000	1.5004	0.7075
1.4994	0.1862	10000	1.2256	0.7532
1.2888	0.2793	15000	1.0994	0.7766
1.1746	0.3723	20000	1.0090	0.7915
1.0994	0.4654	25000	0.9514	0.8021
1.0379	0.5585	30000	0.9029	0.8115
0.9956	0.6516	35000	0.8695	0.8174
0.9647	0.7447	40000	0.8462	0.8216
0.9351	0.8378	45000	0.8274	0.8258
0.9194	0.9309	50000	0.8120	0.8286

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1