metadata

license: mit
tags:
  - generated_from_keras_callback
model-index:
  - name: gpt_image_clef2
    results: []

gpt_image_clef2

This model is a fine-tuned version of gpt2 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': 'AdamWeightDecay', 'learning_rate': {'class_name': 'WarmUp', 'config': {'initial_learning_rate': 0.0005, 'decay_schedule_fn': {'class_name': 'PolynomialDecay', 'config': {'initial_learning_rate': 0.0005, 'decay_steps': 2554800, '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': 0.99, 'epsilon': 0.2, 'amsgrad': False, 'weight_decay_rate': 0.01}
training_precision: float32

Train Loss	Train Rouge	Validation Loss	Validation Rouge	Epoch
1.5255	0.4213	1.0251	0.4284	0
1.1805	0.4673	0.9779	0.4442	1
1.1394	0.4800	0.9561	0.4509	2
1.1168	0.4871	0.9369	0.4595	3
1.1036	0.4915	0.9314	0.4623	4
1.0971	0.4936	0.9283	0.4624	5
1.0946	0.4947	0.9315	0.4617	6
1.0962	0.4947	0.9323	0.4614	7
1.1001	0.4943	0.9405	0.4586	8
1.1065	0.4933	0.9501	0.4560	9
1.1146	0.4913	0.9614	0.4498	10
1.1240	0.4890	0.9726	0.4471	11
1.1341	0.4864	0.9852	0.4429	12
1.1451	0.4836	0.9982	0.4389	13
1.1564	0.4799	1.0160	0.4319	14
1.1680	0.4766	1.0273	0.4296	15
1.1793	0.4732	1.0405	0.4267	16
1.1901	0.4699	1.0556	0.4235	17
1.2007	0.4666	1.0692	0.4184	18
1.2108	0.4632	1.0796	0.4168	19
1.2207	0.4603	1.0998	0.4093	20
1.2299	0.4574	1.1135	0.4057	21
1.2386	0.4547	1.1297	0.4026	22
1.2469	0.4519	1.1396	0.4013	23
1.2540	0.4497	1.1467	0.3960	24
1.2611	0.4475	1.1578	0.3944	25