jobreu/bert-base-cased-hateeval-finetuned

results_trainer

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

• Loss: 0.4417
• Accuracy: 0.799
• F1: 0.7767
• Precision: 0.7296
• Recall: 0.8302

Model description

Test model created as part of an online course on adapters for working with text data.

Intended uses & limitations

This is just a test case for learning.

Training and evaluation data

HateEval 2019 - Task 5 data set

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 1e-05
• train_batch_size: 32
• eval_batch_size: 32
• seed: 108
• 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
• lr_scheduler_warmup_steps: 100
• num_epochs: 3
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1	Precision	Recall
0.4928	1.0	250	0.4792	0.768	0.7089	0.7513	0.6710
0.3599	2.0	500	0.4417	0.799	0.7767	0.7296	0.8302
0.346	3.0	750	0.4399	0.8065	0.7730	0.7636	0.7827

Framework versions

• Transformers 4.52.4
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.21.1