fydhfzh's picture
End of training
dac23fc verified
metadata
license: apache-2.0
base_model: facebook/hubert-base-ls960
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - precision
  - recall
  - f1
model-index:
  - name: hubert-classifier-aug
    results: []

hubert-classifier-aug

This model is a fine-tuned version of facebook/hubert-base-ls960 on an unknown dataset. It achieves the following results on the evaluation set:

  • Loss: 4.4217
  • Accuracy: 0.0135
  • Precision: 0.0002
  • Recall: 0.0135
  • F1: 0.0004
  • Binary: 0.1334

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
  • 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_steps: 500
  • num_epochs: 100
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy Precision Recall F1 Binary
No log 0.19 50 4.4268 0.0135 0.0002 0.0135 0.0004 0.1315
No log 0.38 100 4.4280 0.0135 0.0002 0.0135 0.0004 0.1261
No log 0.58 150 4.4240 0.0135 0.0002 0.0135 0.0004 0.1334
No log 0.77 200 4.4235 0.0135 0.0002 0.0135 0.0004 0.1353
No log 0.96 250 4.4229 0.0135 0.0002 0.0135 0.0004 0.1353
No log 1.15 300 4.4210 0.0162 0.0003 0.0162 0.0005 0.1280
No log 1.34 350 4.4216 0.0162 0.0003 0.0162 0.0005 0.1280
No log 1.53 400 4.4225 0.0135 0.0002 0.0135 0.0004 0.1334
No log 1.73 450 4.4229 0.0135 0.0002 0.0135 0.0004 0.1353
4.4254 1.92 500 4.4217 0.0135 0.0002 0.0135 0.0004 0.1334
4.4254 2.11 550 4.4214 0.0135 0.0002 0.0135 0.0004 0.1334
4.4254 2.3 600 4.4213 0.0135 0.0002 0.0135 0.0004 0.1315
4.4254 2.49 650 4.4213 0.0135 0.0002 0.0135 0.0004 0.1261
4.4254 2.68 700 4.4228 0.0135 0.0002 0.0135 0.0004 0.1334
4.4254 2.88 750 4.4209 0.0135 0.0002 0.0135 0.0004 0.1334
4.4254 3.07 800 4.4210 0.0135 0.0002 0.0135 0.0004 0.1261
4.4254 3.26 850 4.4216 0.0135 0.0002 0.0135 0.0004 0.1315

Framework versions

  • Transformers 4.38.2
  • Pytorch 2.3.0
  • Datasets 2.19.1
  • Tokenizers 0.15.1