metadata

license: mit
tags:
  - generated_from_keras_callback
model-index:
  - name: Regression_xlnet_aug_CustomLoss
    results: []

Regression_xlnet_aug_CustomLoss

This model is a fine-tuned version of xlnet-base-cased 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': 1e-04, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
training_precision: float32

Train Loss	Train Mae	Train Mse	Train R2-score	Validation Loss	Validation Mae	Validation Mse	Validation R2-score	Epoch
0.3976	0.5843	0.4994	0.6170	0.3698	0.6232	0.4576	0.5391	0
0.3938	0.5830	0.4901	0.5860	0.4330	0.6806	0.5537	0.2841	1
0.3822	0.5739	0.4691	-2.6800	0.3660	0.5198	0.4614	0.7579	2
0.3716	0.5723	0.4607	-0.3420	0.3541	0.5461	0.4343	0.7295	3
0.3852	0.5794	0.4793	0.6654	0.3524	0.5654	0.4288	0.6982	4
0.3826	0.5782	0.4787	0.5114	0.3584	0.5991	0.4374	0.6172	5
0.3817	0.5746	0.4754	0.6431	0.3569	0.5945	0.4348	0.6304	6
0.3705	0.5689	0.4541	0.6679	0.3621	0.5210	0.4528	0.7540	7
0.3719	0.5654	0.4567	0.6807	0.3544	0.5446	0.4350	0.7315	8
0.3752	0.5698	0.4638	0.6324	0.3571	0.5343	0.4413	0.7433	9
0.3773	0.5714	0.4655	0.7523	0.3534	0.5787	0.4294	0.6701	10
0.3651	0.5615	0.4490	0.6223	0.3766	0.5372	0.4846	0.7600	11
0.3706	0.5649	0.4529	0.5998	0.3525	0.5683	0.4286	0.6924	12
0.3779	0.5687	0.4619	0.6900	0.3532	0.5771	0.4291	0.6737	13
0.3652	0.5686	0.4521	0.7022	0.3538	0.5478	0.4335	0.7272	14