distilbert-finetuned-tweet-eval

This model is a fine-tuned version of distilbert-base-uncased on an tweet_eval/irony dataset. It achieves the following results on the evaluation set:

• Loss: 0.7944
• Accuracy: 0.6480

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: 9.345135299518317e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 13
• 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	Accuracy
No log	1.0	45	0.6137	0.6649
No log	2.0	90	0.6282	0.6869
No log	3.0	135	0.7045	0.7079

Framework versions

• Transformers 4.38.2
• Pytorch 2.1.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2