tamasheq-99-2

This model is a fine-tuned version of ad019el/tamasheq-99-2 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.3830
• Wer: 0.8701

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: 3e-05
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 32
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 200

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
9.9932	15.79	300	3.5172	1.0
2.9067	31.58	600	1.7973	1.0282
0.7973	47.37	900	1.1744	0.8757
0.4535	63.16	1200	1.2484	0.8475
0.3511	78.95	1500	1.3254	0.8616
0.3156	94.74	1800	1.3830	0.8701

Framework versions

• Transformers 4.32.1
• Pytorch 2.0.1+cu118
• Datasets 2.14.4
• Tokenizers 0.13.3