classification-hate-speech-DE

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

• Loss: 2.6475
• F1 macro: 0.3742
• Weighted: 0.5772
• Balanced accuracy: 0.4995

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: 3e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 11

Training results

Training Loss	Epoch	Step	Validation Loss	F1 macro	Weighted	Balanced accuracy
1.3987	1.0	152	1.1932	0.3716	0.6024	0.4578
0.8176	2.0	304	1.1670	0.4016	0.6136	0.4964
0.6289	3.0	456	1.2036	0.4099	0.6391	0.5131
0.1606	4.0	608	1.7535	0.3725	0.5728	0.5006
0.0285	5.0	760	1.8911	0.4152	0.6312	0.5199
0.0693	6.0	912	2.1700	0.4018	0.6026	0.5087
0.0047	7.0	1064	2.2234	0.3929	0.6128	0.5061
0.0032	8.0	1216	2.5087	0.3759	0.5816	0.4993
0.0025	9.0	1368	2.7507	0.3823	0.5547	0.5173
0.0016	10.0	1520	2.6432	0.3758	0.5808	0.5002
0.0012	11.0	1672	2.6475	0.3742	0.5772	0.4995

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1