Edit model card

hubert-base-ls960-ft

This model is a fine-tuned version of facebook/hubert-base-ls960 on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 0.0123
  • Accuracy: 0.9973

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: 1e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 128
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 5

Training results

Training Loss Epoch Step Accuracy Validation Loss
0.1205 1.0 8558 0.9955 0.0173
0.0638 2.0 17116 0.9973 0.0123
0.0747 3.0 25674 0.9964 0.0183
0.0636 4.0 34232 0.9958 0.0201
0.0531 5.0 42790 0.9967 0.0168

Framework versions

  • Transformers 4.26.1
  • Pytorch 1.11.0+cpu
  • Datasets 2.10.1
  • Tokenizers 0.12.1
Downloads last month
4
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train mazkooleg/0-9up-hubert-base-ls960-ft