TutlaytAI/TrOcr-Hausa

TrOCR-arb-hausaFull

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

• Cer: 0.6361

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-06
• 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
• lr_scheduler_warmup_steps: 100
• num_epochs: 60
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Cer
No log	6.4103	500	4.1384	0.9864
No log	12.8205	1000	3.7169	0.9803
No log	19.2308	1500	3.7846	0.9770
No log	25.6410	2000	3.8778	0.9758

Framework versions

• Transformers 4.53.2
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.2