ghananlpcommunity/w2v-bert-2.0_twi_alpha_v1

<!-- 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. -->

w2v-bert-2.0twialpha_v1

This model is a fine-tuned version of FarmerlineML/w2v-bert-2.0_twi_alpha on an unknown dataset. It achieves the following results on the evaluation set:

• —eval_loss: 0.4134
• —eval_cer: 0.0730
• —eval_wer: 0.2584
• —eval_runtime: 21.7527
• —evalsamplesper_second: 20.457
• —evalstepsper_second: 2.574
• —epoch: 8.5147
• —step: 800

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: 0.0001
• —trainbatchsize: 8
• —evalbatchsize: 8
• —seed: 42
• —gradientaccumulationsteps: 4
• —totaltrainbatch_size: 32
• —optimizer: Use OptimizerNames.ADAMWTORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizerargs=No additional optimizer arguments
• —lrschedulertype: linear
• —lrschedulerwarmup_steps: 400
• —num_epochs: 12
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.52.3
• —Pytorch 2.6.0+cu124
• —Datasets 2.14.4
• —Tokenizers 0.21.1