60  Hyperparameter Tuning with spotpython and PyTorch Lightning Using a CondNet Model

• We use the Diabetes dataset to illustrate the hyperparameter tuning process of a CondNet model using the spotpython package.
• The CondNet model is a conditional neural network that can be used to model conditional distributions [LINK].

Note, the divergence_threshold is set to 5,000, which is based on some pre-experiments with the Diabetes data set.

from spotpython.data.diabetes import Diabetes
from spotpython.hyperdict.light_hyper_dict import LightHyperDict
from spotpython.fun.hyperlight import HyperLight
from spotpython.utils.init import (fun_control_init, surrogate_control_init, design_control_init)
from spotpython.utils.eda import print_exp_table
from spotpython.spot import Spot
from spotpython.utils.file import get_experiment_filename
from math import inf
from spotpython.hyperparameters.values import set_hyperparameter

PREFIX="CondNet_01"

data_set = Diabetes()
input_dim = 10
output_dim = 1
cond_dim = 2

fun_control = fun_control_init(
    PREFIX=PREFIX,
    fun_evals=inf,
    max_time=1,
    data_set = data_set,
    core_model_name="light.regression.NNCondNetRegressor",
    hyperdict=LightHyperDict,
    divergence_threshold=5_000,
    _L_in=input_dim - cond_dim,
    _L_out=1,
    _L_cond=cond_dim,)

fun = HyperLight().fun


set_hyperparameter(fun_control, "optimizer", [ "Adadelta", "Adam", "Adamax"])
set_hyperparameter(fun_control, "l1", [3,4])
set_hyperparameter(fun_control, "epochs", [3,7])
set_hyperparameter(fun_control, "batch_size", [4,5])
set_hyperparameter(fun_control, "dropout_prob", [0.0, 0.025])
set_hyperparameter(fun_control, "patience", [2,3])
set_hyperparameter(fun_control, "lr_mult", [0.1, 20.0])

design_control = design_control_init(init_size=10)

print_exp_table(fun_control)

module_name: light
submodule_name: regression
model_name: NNCondNetRegressor
| name           | type   | default   |   lower |   upper | transform             |
|----------------|--------|-----------|---------|---------|-----------------------|
| l1             | int    | 3         |     3   |   4     | transform_power_2_int |
| epochs         | int    | 4         |     3   |   7     | transform_power_2_int |
| batch_size     | int    | 4         |     4   |   5     | transform_power_2_int |
| act_fn         | factor | ReLU      |     0   |   5     | None                  |
| optimizer      | factor | SGD       |     0   |   2     | None                  |
| dropout_prob   | float  | 0.01      |     0   |   0.025 | None                  |
| lr_mult        | float  | 1.0       |     0.1 |  20     | None                  |
| patience       | int    | 2         |     2   |   3     | transform_power_2_int |
| batch_norm     | factor | 0         |     0   |   1     | None                  |
| initialization | factor | Default   |     0   |   4     | None                  |

spot_tuner = Spot(fun=fun,fun_control=fun_control, design_control=design_control)
res = spot_tuner.run()

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24266.927734375, 'hp_metric': 24266.927734375}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 24001.298828125, 'hp_metric': 24001.298828125}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 21142.806640625, 'hp_metric': 21142.806640625}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 24029.455078125, 'hp_metric': 24029.455078125}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 23382.876953125, 'hp_metric': 23382.876953125}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 24010.201171875, 'hp_metric': 24010.201171875}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23698.623046875, 'hp_metric': 23698.623046875}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 23921.322265625, 'hp_metric': 23921.322265625}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 23699.216796875, 'hp_metric': 23699.216796875}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24038.666015625, 'hp_metric': 24038.666015625}

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23752.283203125, 'hp_metric': 23752.283203125}
spotpython tuning: 21142.806640625 [----------] 0.59%. Success rate: 0.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 22542.3828125, 'hp_metric': 22542.3828125}
spotpython tuning: 21142.806640625 [----------] 1.13%. Success rate: 0.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 21875.419921875, 'hp_metric': 21875.419921875}
spotpython tuning: 21142.806640625 [----------] 2.15%. Success rate: 0.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 22236.15234375, 'hp_metric': 22236.15234375}
spotpython tuning: 21142.806640625 [----------] 3.13%. Success rate: 0.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 16283.1318359375, 'hp_metric': 16283.1318359375}
spotpython tuning: 16283.1318359375 [----------] 4.07%. Success rate: 20.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 22554.375, 'hp_metric': 22554.375}
spotpython tuning: 16283.1318359375 [----------] 4.73%. Success rate: 16.67% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 20658.12890625, 'hp_metric': 20658.12890625}
spotpython tuning: 16283.1318359375 [#---------] 5.47%. Success rate: 14.29% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 8588.830078125, 'hp_metric': 8588.830078125}
spotpython tuning: 8588.830078125 [#---------] 6.03%. Success rate: 25.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23619.22265625, 'hp_metric': 23619.22265625}
spotpython tuning: 8588.830078125 [#---------] 6.63%. Success rate: 22.22% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23825.615234375, 'hp_metric': 23825.615234375}
spotpython tuning: 8588.830078125 [#---------] 7.12%. Success rate: 20.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23993.693359375, 'hp_metric': 23993.693359375}
spotpython tuning: 8588.830078125 [#---------] 7.85%. Success rate: 18.18% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23921.734375, 'hp_metric': 23921.734375}
spotpython tuning: 8588.830078125 [#---------] 8.61%. Success rate: 16.67% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24051.060546875, 'hp_metric': 24051.060546875}
spotpython tuning: 8588.830078125 [#---------] 9.37%. Success rate: 15.38% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23882.166015625, 'hp_metric': 23882.166015625}
spotpython tuning: 8588.830078125 [#---------] 10.06%. Success rate: 14.29% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 22620.6328125, 'hp_metric': 22620.6328125}
spotpython tuning: 8588.830078125 [#---------] 10.78%. Success rate: 13.33% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24091.826171875, 'hp_metric': 24091.826171875}
spotpython tuning: 8588.830078125 [#---------] 11.42%. Success rate: 12.50% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 24000.14453125, 'hp_metric': 24000.14453125}
spotpython tuning: 8588.830078125 [#---------] 12.07%. Success rate: 11.76% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 23777.0234375, 'hp_metric': 23777.0234375}
spotpython tuning: 8588.830078125 [#---------] 12.91%. Success rate: 11.11% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23824.72265625, 'hp_metric': 23824.72265625}
spotpython tuning: 8588.830078125 [#---------] 13.62%. Success rate: 10.53% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 16348.5556640625, 'hp_metric': 16348.5556640625}
spotpython tuning: 8588.830078125 [#---------] 14.99%. Success rate: 10.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 542253.5625, 'hp_metric': 542253.5625}
spotpython tuning: 8588.830078125 [##--------] 15.79%. Success rate: 9.52% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 21935.23046875, 'hp_metric': 21935.23046875}
spotpython tuning: 8588.830078125 [##--------] 18.39%. Success rate: 9.09% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24014.890625, 'hp_metric': 24014.890625}
spotpython tuning: 8588.830078125 [##--------] 20.99%. Success rate: 8.70% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 21583.421875, 'hp_metric': 21583.421875}
spotpython tuning: 8588.830078125 [##--------] 23.72%. Success rate: 8.33% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24219.591796875, 'hp_metric': 24219.591796875}
spotpython tuning: 8588.830078125 [###-------] 26.47%. Success rate: 8.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 23552.76171875, 'hp_metric': 23552.76171875}
spotpython tuning: 8588.830078125 [###-------] 28.70%. Success rate: 7.69% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23819.322265625, 'hp_metric': 23819.322265625}
spotpython tuning: 8588.830078125 [###-------] 29.31%. Success rate: 7.41% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 24035.205078125, 'hp_metric': 24035.205078125}
spotpython tuning: 8588.830078125 [###-------] 30.98%. Success rate: 7.14% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24020.1328125, 'hp_metric': 24020.1328125}
spotpython tuning: 8588.830078125 [####------] 35.31%. Success rate: 6.90% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 23895.53125, 'hp_metric': 23895.53125}
spotpython tuning: 8588.830078125 [####------] 37.66%. Success rate: 6.67% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24009.51171875, 'hp_metric': 24009.51171875}
spotpython tuning: 8588.830078125 [####------] 39.80%. Success rate: 6.45% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 24054.033203125, 'hp_metric': 24054.033203125}
spotpython tuning: 8588.830078125 [####------] 41.62%. Success rate: 6.25% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 24168.806640625, 'hp_metric': 24168.806640625}
spotpython tuning: 8588.830078125 [####------] 44.22%. Success rate: 6.06% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 24035.931640625, 'hp_metric': 24035.931640625}
spotpython tuning: 8588.830078125 [#####-----] 47.13%. Success rate: 5.88% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23980.1328125, 'hp_metric': 23980.1328125}
spotpython tuning: 8588.830078125 [#####-----] 47.95%. Success rate: 5.71% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24006.662109375, 'hp_metric': 24006.662109375}
spotpython tuning: 8588.830078125 [#####-----] 52.16%. Success rate: 5.56% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 23640.08984375, 'hp_metric': 23640.08984375}
spotpython tuning: 8588.830078125 [######----] 55.99%. Success rate: 5.41% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 24108.263671875, 'hp_metric': 24108.263671875}
spotpython tuning: 8588.830078125 [######----] 59.61%. Success rate: 5.26% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 8761.345703125, 'hp_metric': 8761.345703125}
spotpython tuning: 8588.830078125 [######----] 62.02%. Success rate: 5.13% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23936.365234375, 'hp_metric': 23936.365234375}
spotpython tuning: 8588.830078125 [#######---] 66.77%. Success rate: 5.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23658.79296875, 'hp_metric': 23658.79296875}
spotpython tuning: 8588.830078125 [#######---] 67.59%. Success rate: 4.88% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 21879.568359375, 'hp_metric': 21879.568359375}
spotpython tuning: 8588.830078125 [#######---] 68.80%. Success rate: 4.76% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 23994.609375, 'hp_metric': 23994.609375}
spotpython tuning: 8588.830078125 [#######---] 70.74%. Success rate: 4.65% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 22487.458984375, 'hp_metric': 22487.458984375}
spotpython tuning: 8588.830078125 [#######---] 73.44%. Success rate: 4.55% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 22477.447265625, 'hp_metric': 22477.447265625}
spotpython tuning: 8588.830078125 [#######---] 74.03%. Success rate: 4.44% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 13692.08984375, 'hp_metric': 13692.08984375}
spotpython tuning: 8588.830078125 [########--] 78.93%. Success rate: 4.35% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24043.455078125, 'hp_metric': 24043.455078125}
spotpython tuning: 8588.830078125 [########--] 81.04%. Success rate: 4.26% 

Using spacefilling design as fallback.

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 2.6 K │ [[16, 8], [16, 2]] │   [16, 8] │
│ 1 │ layers     │ Sequential       │    153 │ train │ 4.2 K │            [16, 8] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 249                                                                                              
Non-trainable params: 0                                                                                            
Total params: 249                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 18                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 6.7 K                                                                                                 

train_model result: {'val_loss': 23562.740234375, 'hp_metric': 23562.740234375}
spotpython tuning: 8588.830078125 [########--] 81.62%. Success rate: 4.17% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23973.388671875, 'hp_metric': 23973.388671875}
spotpython tuning: 8588.830078125 [########--] 84.28%. Success rate: 4.08% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 23741.322265625, 'hp_metric': 23741.322265625}
spotpython tuning: 8588.830078125 [#########-] 88.13%. Success rate: 4.00% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 52557.8828125, 'hp_metric': 52557.8828125}
spotpython tuning: 8588.830078125 [#########-] 91.41%. Success rate: 3.92% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 23981.197265625, 'hp_metric': 23981.197265625}
spotpython tuning: 8588.830078125 [#########-] 92.16%. Success rate: 3.85% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │ 10.2 K │ [[32, 8], [32, 2]] │  [32, 16] │
│ 1 │ layers     │ Sequential       │    691 │ train │ 34.2 K │           [32, 16] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 883                                                                                              
Non-trainable params: 0                                                                                            
Total params: 883                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 36                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 44.4 K                                                                                                

train_model result: {'val_loss': 24007.826171875, 'hp_metric': 24007.826171875}
spotpython tuning: 8588.830078125 [#########-] 93.96%. Success rate: 3.77% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 24064.419921875, 'hp_metric': 24064.419921875}
spotpython tuning: 8588.830078125 [##########] 95.29%. Success rate: 3.70% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 5181.326171875, 'hp_metric': 5181.326171875}
spotpython tuning: 5181.326171875 [##########] 96.04%. Success rate: 5.45% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃ FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │     96 │ train │ 5.1 K │ [[32, 8], [32, 2]] │   [32, 8] │
│ 1 │ layers     │ Sequential       │    197 │ train │ 8.3 K │            [32, 8] │   [32, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴───────┴────────────────────┴───────────┘

Trainable params: 293                                                                                              
Non-trainable params: 0                                                                                            
Total params: 293                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 24                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 13.4 K                                                                                                

train_model result: {'val_loss': 20914.025390625, 'hp_metric': 20914.025390625}
spotpython tuning: 5181.326171875 [##########] 99.37%. Success rate: 5.36% 

┏━━━┳━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━┳━━━━━━━━┳━━━━━━━┳━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━┓
┃   ┃ Name       ┃ Type             ┃ Params ┃ Mode  ┃  FLOPs ┃           In sizes ┃ Out sizes ┃
┡━━━╇━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━╇━━━━━━━━╇━━━━━━━╇━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━┩
│ 0 │ cond_layer │ ConditionalLayer │    192 │ train │  5.1 K │ [[16, 8], [16, 2]] │  [16, 16] │
│ 1 │ layers     │ Sequential       │    587 │ train │ 17.1 K │           [16, 16] │   [16, 1] │
└───┴────────────┴──────────────────┴────────┴───────┴────────┴────────────────────┴───────────┘

Trainable params: 779                                                                                              
Non-trainable params: 0                                                                                            
Total params: 779                                                                                                  
Total estimated model params size (MB): 0                                                                          
Modules in train mode: 26                                                                                          
Modules in eval mode: 0                                                                                            
Total FLOPs: 22.2 K                                                                                                

train_model result: {'val_loss': 182654.625, 'hp_metric': 182654.625}
spotpython tuning: 5181.326171875 [##########] 100.00%. Success rate: 5.26% Done...

Experiment saved to CondNet_01_res.pkl

60.1 Looking at the Results

60.1.1 Tuning Progress

After the hyperparameter tuning run is finished, the progress of the hyperparameter tuning can be visualized with spotpython’s method plot_progress. The black points represent the performace values (score or metric) of hyperparameter configurations from the initial design, whereas the red points represents the hyperparameter configurations found by the surrogate model based optimization.

spot_tuner.plot_progress()

60.1.2 Tuned Hyperparameters and Their Importance

Results can be printed in tabular form.

from spotpython.utils.eda import print_res_table
print_res_table(spot_tuner)

| name           | type   | default   |   lower |   upper | tuned   | transform             |   importance | stars   |
|----------------|--------|-----------|---------|---------|---------|-----------------------|--------------|---------|
| l1             | int    | 3         |     3.0 |     4.0 | 4.0     | transform_power_2_int |         0.01 |         |
| epochs         | int    | 4         |     3.0 |     7.0 | 7.0     | transform_power_2_int |         0.01 |         |
| batch_size     | int    | 4         |     4.0 |     5.0 | 4.0     | transform_power_2_int |       100.00 | ***     |
| act_fn         | factor | ReLU      |     0.0 |     5.0 | ReLU    | None                  |         0.01 |         |
| optimizer      | factor | SGD       |     0.0 |     2.0 | Adam    | None                  |        12.24 | *       |
| dropout_prob   | float  | 0.01      |     0.0 |   0.025 | 0.0     | None                  |         0.01 |         |
| lr_mult        | float  | 1.0       |     0.1 |    20.0 | 20.0    | None                  |         0.01 |         |
| patience       | int    | 2         |     2.0 |     3.0 | 2.0     | transform_power_2_int |         0.01 |         |
| batch_norm     | factor | 0         |     0.0 |     1.0 | 0       | None                  |         0.01 |         |
| initialization | factor | Default   |     0.0 |     4.0 | Default | None                  |         0.01 |         |

A histogram can be used to visualize the most important hyperparameters.

spot_tuner.plot_importance(threshold=1.0)

spot_tuner.plot_important_hyperparameter_contour(max_imp=3)

l1:  0.005952705344705705
epochs:  0.005952705344705705
batch_size:  100.0
act_fn:  0.005952705344705705
optimizer:  12.238623055489287
dropout_prob:  0.005952705344705705
lr_mult:  0.005952705344705705
patience:  0.005952705344705705
batch_norm:  0.005952705344705705
initialization:  0.005952705344705705

60.1.3 Get the Tuned Architecture

import pprint
from spotpython.hyperparameters.values import get_tuned_architecture
config = get_tuned_architecture(spot_tuner)
pprint.pprint(config)

{'act_fn': ReLU(),
 'batch_norm': False,
 'batch_size': 16,
 'dropout_prob': 0.0,
 'epochs': 128,
 'initialization': 'Default',
 'l1': 16,
 'lr_mult': 20.0,
 'optimizer': 'Adam',
 'patience': 4}