testing_pretrained_tamasheq_only

This model is a fine-tuned version of YassineBenlaria/testing_pretrained_tamasheq_only on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.4042
• Wer: 0.8037

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: 350

Training results

Training Loss	Epoch	Step	Validation Loss	Wer
3.6733	35.29	300	2.7902	1.0
2.3267	70.59	600	1.7481	0.8778
1.003	105.88	900	1.8251	0.8
0.6905	141.18	1200	2.0436	0.8148
0.5074	176.47	1500	2.1611	0.8259
0.3883	211.76	1800	2.2562	0.8333
0.3431	247.06	2100	2.3761	0.8148
0.3053	282.35	2400	2.3995	0.8074
0.2898	317.65	2700	2.4042	0.8037

Framework versions

• Transformers 4.33.2
• Pytorch 2.0.1+cu118
• Datasets 2.14.5
• Tokenizers 0.13.3