amanneo/mail-generator-mini-v2

This model was trained from scratch on an unknown dataset. It achieves the following results on the evaluation set:

• Train Loss: 0.5212
• Train Accuracy: 0.0027
• Validation Loss: 5.5781
• Validation Accuracy: 0.0
• Epoch: 99

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: {'inner_optimizer': {'class_name': 'AdamWeightDecay', 'config': {'name': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 5e-05, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 5e-05, 'decay_steps': -994, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'passive_serialization': True}, 'warmup_steps': 1000, 'power': 1.0, 'name': None}}, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False, 'weight_decay_rate': 0.01}}, 'dynamic': True, 'initial_scale': 32768.0, 'dynamic_growth_steps': 2000}
• training_precision: mixed_float16

Training results

Train Loss	Train Accuracy	Validation Loss	Validation Accuracy	Epoch
2.5928	0.0171	5.5430	0.0048	0
2.6003	0.0207	5.5430	0.0048	1
2.5954	0.0171	5.5508	0.0048	2
2.5775	0.0190	5.5508	0.0024	3
2.5758	0.0231	5.5508	0.0024	4
2.5742	0.0207	5.5586	0.0048	5
2.5547	0.0209	5.5586	0.0048	6
2.5566	0.0188	5.5586	0.0048	7
2.5391	0.0193	5.5586	0.0048	8
2.5378	0.0215	5.5508	0.0048	9
2.5238	0.0188	5.5469	0.0048	10
2.5150	0.0160	5.5508	0.0048	11
2.4967	0.0174	5.5508	0.0071	12
2.4691	0.0193	5.5430	0.0071	13
2.4626	0.0163	5.5430	0.0071	14
2.4417	0.0231	5.5352	0.0048	15
2.4323	0.0215	5.5352	0.0048	16
2.4193	0.0226	5.5469	0.0048	17
2.4170	0.0185	5.5469	0.0048	18
2.3743	0.0193	5.5312	0.0048	19
2.3730	0.0207	5.5312	0.0048	20
2.3535	0.0198	5.5312	0.0048	21
2.3372	0.0182	5.5312	0.0071	22
2.3324	0.0177	5.5312	0.0048	23
2.3011	0.0204	5.5195	0.0048	24
2.2650	0.0212	5.5117	0.0048	25
2.2568	0.0198	5.5078	0.0048	26
2.2331	0.0196	5.5156	0.0048	27
2.2021	0.0193	5.5078	0.0048	28
2.1807	0.0204	5.5039	0.0048	29
2.1691	0.0190	5.5	0.0	30
2.1463	0.0174	5.4766	0.0	31
2.1097	0.0196	5.4844	0.0	32
2.1014	0.0179	5.4844	0.0024	33
2.0833	0.0177	5.4844	0.0024	34
2.0423	0.0201	5.4844	0.0	35
2.0163	0.0198	5.4844	0.0	36
1.9909	0.0168	5.4883	0.0	37
1.9774	0.0207	5.4805	0.0	38
1.9414	0.0207	5.4844	0.0	39
1.9206	0.0215	5.4766	0.0	40
1.8849	0.0182	5.4805	0.0	41
1.8732	0.0193	5.4648	0.0	42
1.8460	0.0160	5.4609	0.0	43
1.8171	0.0168	5.4648	0.0	44
1.7791	0.0201	5.4531	0.0	45
1.7583	0.0158	5.4570	0.0	46
1.7360	0.0171	5.4570	0.0	47
1.7061	0.0120	5.4297	0.0	48
1.6802	0.0155	5.4258	0.0	49
1.6551	0.0182	5.4141	0.0	50
1.6289	0.0130	5.4219	0.0	51
1.5981	0.0130	5.3945	0.0	52
1.5656	0.0128	5.4297	0.0	53
1.5535	0.0168	5.4219	0.0	54
1.5184	0.0141	5.4102	0.0	55
1.4943	0.0149	5.4023	0.0	56
1.4616	0.0122	5.4062	0.0	57
1.4344	0.0111	5.4062	0.0	58
1.3965	0.0111	5.4141	0.0	59
1.3643	0.0122	5.4375	0.0	60
1.3309	0.0087	5.4453	0.0	61
1.3215	0.0090	5.4648	0.0	62
1.3058	0.0084	5.4727	0.0	63
1.2700	0.0109	5.4453	0.0	64
1.2396	0.0079	5.4609	0.0	65
1.2189	0.0092	5.4375	0.0	66
1.1855	0.0079	5.4375	0.0	67
1.1592	0.0073	5.4375	0.0	68
1.1219	0.0071	5.4648	0.0	69
1.1071	0.0065	5.4570	0.0	70
1.0848	0.0060	5.4375	0.0	71
1.0581	0.0076	5.4453	0.0	72
1.0316	0.0090	5.4570	0.0	73
1.0068	0.0063	5.4219	0.0	74
0.9832	0.0060	5.4570	0.0	75
0.9534	0.0046	5.4570	0.0	76
0.9378	0.0057	5.4648	0.0	77
0.9170	0.0033	5.4844	0.0	78
0.8941	0.0041	5.4883	0.0	79
0.8666	0.0030	5.4922	0.0	80
0.8419	0.0054	5.4375	0.0	81
0.8200	0.0035	5.4492	0.0	82
0.8020	0.0022	5.4648	0.0	83
0.7785	0.0057	5.4883	0.0	84
0.7607	0.0052	5.4648	0.0	85
0.7454	0.0041	5.5078	0.0	86
0.7208	0.0024	5.5078	0.0	87
0.7040	0.0027	5.5078	0.0	88
0.6799	0.0041	5.5156	0.0	89
0.6594	0.0030	5.5312	0.0	90
0.6397	0.0030	5.5312	0.0	91
0.6217	0.0030	5.5195	0.0	92
0.6112	0.0033	5.5195	0.0	93
0.5937	0.0046	5.5625	0.0	94
0.5745	0.0035	5.5625	0.0	95
0.5616	0.0027	5.5586	0.0	96
0.5468	0.0043	5.5742	0.0	97
0.5354	0.0027	5.5781	0.0	98
0.5212	0.0027	5.5781	0.0	99

Framework versions

• Transformers 4.23.1
• TensorFlow 2.9.2
• Datasets 2.6.1
• Tokenizers 0.13.1