twitter_complaints_model

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

• Loss: 1.1767
• Accuracy: 0.7143

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: 1e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 100

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6341	5.0	20	0.6401	0.6425
0.5406	10.0	40	0.5885	0.6625
0.3126	15.0	60	0.5854	0.7096
0.111	20.0	80	0.7057	0.7132
0.0418	25.0	100	0.8403	0.7102
0.0234	30.0	120	0.9430	0.7070
0.0194	35.0	140	0.9413	0.7152
0.0137	40.0	160	0.9818	0.7152
0.0106	45.0	180	1.0088	0.7167
0.0088	50.0	200	1.0585	0.7134
0.0083	55.0	220	1.0737	0.7143
0.0076	60.0	240	1.1009	0.7134
0.0067	65.0	260	1.0975	0.7187
0.0059	70.0	280	1.1262	0.7140
0.0057	75.0	300	1.1484	0.7134
0.005	80.0	320	1.1608	0.7137
0.0046	85.0	340	1.1709	0.7134
0.0045	90.0	360	1.1762	0.7137
0.0049	95.0	380	1.1752	0.7143
0.0045	100.0	400	1.1767	0.7143

Framework versions

• Transformers 4.53.3
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.2