wdevinsp/indobert-base-uncased-finetuned-digestive-qna

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

Model description

More information needed

More information needed

More information needed

The following hyperparameters were used during training:

optimizer: {'name': 'Adam', 'weight_decay': None, 'clipnorm': None, 'global_clipnorm': None, 'clipvalue': None, 'use_ema': False, 'ema_momentum': 0.99, 'ema_overwrite_frequency': None, 'jit_compile': True, 'is_legacy_optimizer': False, 'learning_rate': {'module': 'keras.optimizers.schedules', 'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 2e-05, 'decay_steps': 3700, 'end_learning_rate': 0.0, 'power': 1.0, 'cycle': False, 'name': None}, 'registered_name': None}, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-08, 'amsgrad': False}
training_precision: float32

Train Loss	Train End Logits Accuracy	Train Start Logits Accuracy	Epoch
3.1872	0.3564	0.3733	0
2.1076	0.4189	0.4088	1
1.6782	0.4443	0.5507	2
1.1711	0.5507	0.7111	3
0.6231	0.7720	0.8429	4
0.4296	0.8463	0.8902	5
0.2544	0.9189	0.9409	6
0.1853	0.9358	0.9510	7
0.1411	0.9544	0.9578	8
0.1215	0.9493	0.9662	9
0.1141	0.9645	0.9628	10
0.1195	0.9561	0.9696	11
0.0851	0.9662	0.9645	12
0.1255	0.9527	0.9493	13
0.0897	0.9595	0.9730	14
0.0860	0.9578	0.9696	15
0.0710	0.9595	0.9713	16
0.0669	0.9628	0.9713	17
0.0634	0.9797	0.9730	18
0.0765	0.9662	0.9814	19
0.0732	0.9679	0.9730	20
0.0586	0.9696	0.9713	21
0.0572	0.9679	0.9764	22
0.0518	0.9747	0.9764	23
0.0500	0.9713	0.9764	24
0.0464	0.9696	0.9713	25
0.0470	0.9814	0.9730	26
0.0575	0.9797	0.9764	27
0.0660	0.9679	0.9696	28
0.0576	0.9713	0.9696	29
0.0515	0.9696	0.9747	30
0.0547	0.9780	0.9696	31
0.0430	0.9730	0.9780	32
0.0408	0.9814	0.9747	33
0.0475	0.9747	0.9747	34
0.0536	0.9764	0.9797	35
0.0494	0.9696	0.9730	36
0.0391	0.9814	0.9814	37