shawhin's picture
shawhin/distilbert-base-uncased-lora-text-classification
bd902e3
metadata
license: apache-2.0
base_model: distilbert-base-uncased
tags:
  - generated_from_trainer
metrics:
  - accuracy
model-index:
  - name: distilbert-base-uncased-lora-text-classification
    results: []

distilbert-base-uncased-lora-text-classification

This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:

  • Loss: 1.0684
  • Accuracy: {'accuracy': 0.879}

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: 0.001
  • train_batch_size: 4
  • eval_batch_size: 4
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
No log 1.0 250 0.4266 {'accuracy': 0.87}
0.4232 2.0 500 0.4260 {'accuracy': 0.88}
0.4232 3.0 750 0.5071 {'accuracy': 0.885}
0.2213 4.0 1000 0.7424 {'accuracy': 0.875}
0.2213 5.0 1250 0.7885 {'accuracy': 0.881}
0.067 6.0 1500 0.9312 {'accuracy': 0.872}
0.067 7.0 1750 0.9669 {'accuracy': 0.874}
0.0238 8.0 2000 1.0856 {'accuracy': 0.874}
0.0238 9.0 2250 1.0637 {'accuracy': 0.88}
0.0066 10.0 2500 1.0684 {'accuracy': 0.879}

Framework versions

  • Transformers 4.32.1
  • Pytorch 2.0.1
  • Datasets 2.14.4
  • Tokenizers 0.13.2