hamsa-beta-v0.3Q

This model is a fine-tuned version of nadsoft/hamsa-v0.1-beta on the nadsoft/arabic-98 dataset. It achieves the following results on the evaluation set:

• Loss: 0.2362
• Wer Ortho: 21.12
• Wer: 19.3029

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: 1e-05
• train_batch_size: 32
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• training_steps: 4000
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Wer Ortho	Wer
0.2617	0.25	1000	0.2684	22.16	18.8134
0.227	0.5	2000	0.2565	18.6971	16.7482
0.2585	0.75	3000	0.2442	18.2400	16.3304
0.2632	1.0	4000	0.2362	21.12	19.3029

Framework versions

• Transformers 4.36.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0