Samuael/character_based_tigre_233

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

characterbasedtigre_233

This model is a fine-tuned version of Samuael/asr-amharic-phoneme-based-233 on the alffa_amharic dataset. It achieves the following results on the evaluation set:

• —eval_loss: 1.0735
• —eval_wer: 0.6463
• —eval_cer: 0.2341
• —eval_runtime: 14.3391
• —evalsamplesper_second: 6.556
• —evalstepsper_second: 0.837
• —epoch: 5.6
• —step: 140

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: 0.0005
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 1
• —num_epochs: 10
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.38.2
• —Pytorch 2.2.1+cu121
• —Datasets 2.18.0
• —Tokenizers 0.15.2