README.md

metadata

library_name: transformers
base_model: monologg/koelectra-small-finetuned-sentiment
tags:
  - generated_from_trainer
metrics:
  - accuracy
  - f1
model-index:
  - name: koelectra_emotion_v2
    results: []

koelectra_emotion_v2

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: 1.4260
• Accuracy: 0.6118
• F1: 0.6172

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
1.5444	1.0	601	1.3725	0.4626	0.4567
1.218	2.0	1202	1.2090	0.5664	0.5693
1.0556	3.0	1803	1.1186	0.6031	0.6075
0.9497	4.0	2404	1.1470	0.5981	0.6056
0.8734	5.0	3005	1.1450	0.6005	0.6067
0.808	6.0	3606	1.1479	0.6136	0.6181
0.7489	7.0	4207	1.1225	0.6287	0.6348
0.6926	8.0	4808	1.2075	0.6096	0.6177
0.6472	9.0	5409	1.2047	0.6180	0.6227
0.6028	10.0	6010	1.2248	0.6194	0.6249
0.5624	11.0	6611	1.2474	0.6154	0.6215
0.5303	12.0	7212	1.2627	0.6203	0.6257
0.4956	13.0	7813	1.2977	0.6191	0.6245
0.4662	14.0	8414	1.3655	0.6081	0.6144
0.4439	15.0	9015	1.3801	0.6067	0.6124
0.4221	16.0	9616	1.3854	0.6124	0.6167
0.4097	17.0	10217	1.4101	0.6105	0.6164
0.3921	18.0	10818	1.4359	0.6054	0.6109
0.3799	19.0	11419	1.4269	0.6111	0.6169
0.369	20.0	12020	1.4260	0.6118	0.6172

Framework versions

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