metadata

license: apache-2.0
tags:
  - generated_from_trainer
datasets:
  - clinc_oos
model-index:
  - name: distilbert-base-uncased-finetuned-clinc
    results: []

distilbert-base-uncased-finetuned-clinc

This model is a fine-tuned version of distilbert-base-uncased on the clinc_oos dataset.

Model description

This is an initial example of knowledge-distillation where the student loss is all cross-entropy loss $L_{CE}$ of the ground-truth labels and none of the distillation loss.

Intended uses & limitations

More information needed

Training and evaluation data

The training and evaluation data come straight from the train and validation splits in the clinc_oos dataset, respectively; and tokenized using the distilbert-base-uncased tokenization.

Training procedure

Please see page 224 in Chapter 8: Making Transformers Efficient in Production, Natural Language Processing with Transformers, May 2022.

Training hyperparameters

The following hyperparameters were used during training:

alpha: 1.0
temperature: 2.0
learning_rate: 2e-05
train_batch_size: 48
eval_batch_size: 48
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 5

Framework versions

Transformers 4.16.2
Pytorch 2.1.2+cu121
Datasets 1.16.1
Tokenizers 0.15.1