jysh1023/tiny-bert-sst2-distilled_qat

<!-- 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. -->

tiny-bert-sst2-distilled_qat

This model is a fine-tuned version of jysh1023/tiny-bert-sst2-distilled_qat_test on the glue dataset. It achieves the following results on the evaluation set:

• —eval_loss: 0.6176
• —eval_accuracy: 0.7649
• —eval_runtime: 0.1617
• —evalsamplesper_second: 5391.756
• —evalstepsper_second: 43.282
• —epoch: 1.0
• —step: 527

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: 1.8904711598645274e-05
• —trainbatchsize: 128
• —evalbatchsize: 128
• —seed: 33
• —optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• —lrschedulertype: linear
• —num_epochs: 9
• —mixedprecisiontraining: Native AMP

Framework versions

• —Transformers 4.35.2
• —Pytorch 2.1.0+cu118
• —Datasets 2.15.0
• —Tokenizers 0.15.0