madatnlp/gamza-bart-for-kormath128

This model is a fine-tuned version of gogamza/kobart-base-v2 on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.1429
• Validation Loss: 0.3575
• Epoch: 42

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:

• optimizer: {'name': 'Adam', 'learning_rate': 1e-04, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
• training_precision: float32

Training results

Train Loss	Validation Loss	Epoch
5.9513	3.2241	0
2.6808	1.8567	1
1.6770	1.2966	2
1.2253	1.0402	3
1.0279	0.9159	4
0.9241	0.8158	5
0.8570	0.8047	6
0.8130	0.7684	7
0.7771	0.7817	8
0.7522	0.7653	9
0.7318	0.6813	10
0.7111	0.6535	11
0.6916	0.6719	12
0.6901	0.7191	13
0.6551	0.6330	14
0.6495	0.6242	15
0.6258	0.6048	16
0.6184	0.6590	17
0.6055	0.6622	18
0.5946	0.6377	19
0.5807	0.5994	20
0.5781	0.5797	21
0.5644	0.6154	22
0.5466	0.5777	23
0.5417	0.6324	24
0.5204	0.5763	25
0.5081	0.5751	26
0.4923	0.5908	27
0.4616	0.5433	28
0.4238	0.4823	29
0.3765	0.4474	30
0.3447	0.4306	31
0.3156	0.3817	32
0.2832	0.3824	33
0.2632	0.3204	34
0.2365	0.3539	35
0.2179	0.3162	36
0.2024	0.3385	37
0.1860	0.3367	38
0.1801	0.3019	39
0.1629	0.3045	40
0.1533	0.2567	41
0.1429	0.3575	42

Framework versions

• Transformers 4.18.0
• TensorFlow 2.8.0
• Datasets 2.1.0
• Tokenizers 0.12.1