koelectra_emotion_v2_2

This model is a fine-tuned version of monologg/koelectra-small-finetuned-sentiment on the None dataset. It achieves the following results on the evaluation set:

• Loss: 1.4392
• Accuracy: 0.5249
• F1: 0.5227

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.8251	1.0	601	1.7175	0.3130	0.2585
1.6569	2.0	1202	1.5616	0.4007	0.3821
1.5086	3.0	1803	1.4698	0.4397	0.4258
1.4099	4.0	2404	1.4268	0.4661	0.4592
1.3318	5.0	3005	1.3885	0.4779	0.4770
1.2688	6.0	3606	1.3679	0.4927	0.4849
1.2213	7.0	4207	1.3497	0.5071	0.5015
1.1756	8.0	4808	1.3405	0.5144	0.5099
1.1328	9.0	5409	1.3548	0.5162	0.5125
1.0964	10.0	6010	1.3864	0.5151	0.5081
1.0675	11.0	6611	1.3606	0.5198	0.5162
1.0368	12.0	7212	1.3719	0.5195	0.5170
1.0044	13.0	7813	1.3695	0.5256	0.5222
0.9808	14.0	8414	1.3896	0.5216	0.5196
0.9554	15.0	9015	1.4033	0.5228	0.5192
0.9307	16.0	9616	1.4135	0.5247	0.5236
0.9131	17.0	10217	1.4193	0.5198	0.5181
0.8969	18.0	10818	1.4259	0.5276	0.5243
0.8825	19.0	11419	1.4385	0.5245	0.5224
0.8779	20.0	12020	1.4392	0.5249	0.5227

Framework versions

• Transformers 4.56.1
• Pytorch 2.8.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.0