humor-no-humor / README.md
mohameddhiab's picture
Update README.md
cec42ea
metadata
license: apache-2.0
tags:
  - generated_from_trainer
metrics:
  - f1
model-index:
  - name: humor-no-humor
    results: []

humor-no-humor

This model is a fine-tuned version of distilbert-base-uncased on a joke/no-joke dataset in order to detect humor. It achieves the following results on the evaluation set:

  • Loss: 0.1269
  • F1: 0.9537

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: 2e-06
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss F1
0.1707 1.0 1677 0.1398 0.9423
0.1427 2.0 3354 0.1291 0.9531
0.1384 3.0 5031 0.1269 0.9537

Framework versions

  • Transformers 4.31.0.dev0
  • Pytorch 2.0.0
  • Datasets 2.1.0
  • Tokenizers 0.13.3