edyfjm07/distilbert-base-uncased-QA2-finetuned-squad-es

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

• Train Loss: 0.0138
• Train End Logits Accuracy: 0.9947
• Train Start Logits Accuracy: 1.0
• Validation Loss: 1.7511
• Validation End Logits Accuracy: 0.7931
• Validation Start Logits Accuracy: 0.7994
• Epoch: 45

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', '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': 0.0001, 'decay_steps': 5474, '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

Training results

Train Loss	Train End Logits Accuracy	Train Start Logits Accuracy	Validation Loss	Validation End Logits Accuracy	Validation Start Logits Accuracy	Epoch
2.3428	0.4160	0.4317	1.3438	0.5611	0.6458	0
1.1526	0.6261	0.6397	1.0597	0.6677	0.7429	1
0.7612	0.7269	0.7647	1.0245	0.7210	0.7806	2
0.5528	0.7836	0.8319	1.2436	0.7116	0.7712	3
0.4667	0.8340	0.8435	1.0705	0.7524	0.7555	4
0.3834	0.8813	0.8687	1.1209	0.7586	0.7712	5
0.3678	0.8634	0.8876	1.2341	0.7618	0.7649	6
0.2555	0.9044	0.9181	1.1561	0.7649	0.8056	7
0.2151	0.9160	0.9328	1.0908	0.7931	0.7994	8
0.1855	0.9286	0.9475	1.2809	0.7994	0.7774	9
0.1654	0.9443	0.9454	1.3974	0.7837	0.7806	10
0.1282	0.9464	0.9517	1.4260	0.7774	0.7837	11
0.1313	0.9443	0.9601	1.4537	0.7900	0.7962	12
0.1301	0.9517	0.9590	1.1851	0.7774	0.8150	13
0.1089	0.9548	0.9590	1.2442	0.7774	0.8088	14
0.1023	0.9601	0.9622	1.4575	0.7931	0.7931	15
0.0956	0.9590	0.9685	1.5160	0.7837	0.7900	16
0.0712	0.9727	0.9737	1.5741	0.7900	0.8088	17
0.0752	0.9674	0.9790	1.4401	0.7931	0.7994	18
0.0604	0.9737	0.9779	1.6410	0.7962	0.8088	19
0.0497	0.9758	0.9821	1.5655	0.7962	0.8119	20
0.0668	0.9685	0.9811	1.3480	0.7806	0.7962	21
0.0567	0.9769	0.9800	1.3820	0.7900	0.8088	22
0.0550	0.9769	0.9832	1.3593	0.7806	0.8056	23
0.0399	0.9821	0.9884	1.5254	0.7868	0.7931	24
0.0320	0.9842	0.9874	1.5801	0.7868	0.7994	25
0.0296	0.9832	0.9884	1.6310	0.7962	0.7962	26
0.0307	0.9863	0.9926	1.4756	0.7774	0.7900	27
0.0254	0.9863	0.9895	1.7564	0.7774	0.7931	28
0.0255	0.9853	0.9937	1.6061	0.7774	0.7962	29
0.0214	0.9863	0.9937	1.7697	0.7712	0.8056	30
0.0283	0.9842	0.9863	1.8398	0.7806	0.7900	31
0.0182	0.9905	0.9926	1.8756	0.7837	0.7994	32
0.0252	0.9832	0.9947	1.8182	0.7837	0.7962	33
0.0222	0.9863	0.9947	1.7854	0.7837	0.7931	34
0.0216	0.9884	0.9947	1.5707	0.7931	0.8025	35
0.0161	0.9937	0.9916	1.7071	0.7806	0.8025	36
0.0146	0.9926	0.9926	1.7827	0.7868	0.7962	37
0.0148	0.9905	0.9947	1.8678	0.7868	0.7931	38
0.0117	0.9884	0.9968	1.7944	0.7868	0.7900	39
0.0137	0.9905	0.9958	1.7666	0.7900	0.7931	40
0.0160	0.9874	0.9958	1.7644	0.7868	0.7962	41
0.0150	0.9916	0.9937	1.7783	0.7868	0.8025	42
0.0128	0.9895	0.9958	1.7480	0.7900	0.7994	43
0.0102	0.9937	0.9947	1.7432	0.7931	0.7994	44
0.0138	0.9947	1.0	1.7511	0.7931	0.7994	45

Framework versions

• Transformers 4.41.2
• TensorFlow 2.15.0
• Datasets 2.20.0
• Tokenizers 0.19.1