bertweet-large-sexism-detector

This model is a fine-tuned version of NLP-LTU/bertweet-large-sexism-detector on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.9648
• Icm: 0.2479
• Icmnorm: 0.6258
• Fmeasure: 0.7526

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-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• distributed_type: multi-GPU
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 10
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Icm	Icmnorm	Fmeasure
0.8774	1.0	771	1.1253	-0.1346	0.4317	0.5634
0.9551	2.0	1542	0.8275	0.1264	0.5641	0.7110
0.9559	3.0	2313	0.9648	0.2479	0.6258	0.7526
0.6926	4.0	3084	1.5632	0.1570	0.5797	0.7172
0.4547	5.0	3855	1.8028	0.1284	0.5652	0.7098
0.2611	6.0	4626	1.9528	0.2025	0.6027	0.7359
0.1528	7.0	5397	2.1400	0.1119	0.5568	0.7073
0.1173	8.0	6168	2.1909	0.1524	0.5773	0.7195
0.1096	9.0	6939	2.4630	0.1166	0.5591	0.7073
0.0535	10.0	7710	2.4917	0.1809	0.5918	0.7276

Framework versions

• Transformers 4.38.2
• Pytorch 2.3.0+cu121
• Datasets 2.18.0
• Tokenizers 0.15.2