modelCategory = NN
model_id = e2e_nn_micro_v2
modelName = mlp_small

==================================================
Architecture
==================================================
Genome = [[[[10.0, 1.0], [11.0, 1.0]], [[2.0, 0.0], [1.0, 1.0]], [[11.0, 3.0], [11.0, 3.0]], [[4.0, 1.0], [2.0, 3.0731720145149364]], [[3.0, 5.0], [6.0, 0.0]]], [[[10.0, 1.0], [10.0, 0.0]], [[0.0, 2.0], [3.0, 1.0]], [[11.0, 2.0], [0.0, 1.0]], [[8.0, 3.0], [10.0, 2.0]], [[11.0, 4.0], [9.0, 0.0]]]]
Genotype = Genotype(normal=[('batch_norm', 1), ('dropout_03', 1), ('linear_16', 0), ('skip_connect', 1), ('dropout_03', 3), ('dropout_03', 3), ('linear_64', 1), ('linear_16', 3), ('linear_32', 5), ('linear_256', 0)], normal_concat=[2, 4, 6], reduce=[('batch_norm', 1), ('batch_norm', 0), ('none', 2), ('linear_32', 1), ('dropout_03', 2), ('none', 1), ('gated_linear', 3), ('batch_norm', 2), ('dropout_03', 4), ('feature_cross', 0)], reduce_concat=[5, 6])
param size = 1.14430MB
total params = 299971
flops = 0.2991M
valid_acc = 89.58

==================================================
Model Configuration
==================================================
search_space = micro
model_config_id = mlp_small
task_type = classification
input_dim = 4
output_dim = 3
C (channels) = 32
layers = 4
epochs = 10

==================================================
Training Progress
==================================================
Epoch   1 | Train Loss: 0.889601 | Val Loss: 0.482999 | LR: 0.009755
Epoch   2 | Train Loss: 0.215365 | Val Loss: 0.048737 | LR: 0.009045
Epoch   3 | Train Loss: 0.315507 | Val Loss: 0.022136 | LR: 0.007939
Epoch   4 | Train Loss: 0.688010 | Val Loss: 0.035093 | LR: 0.006545
Epoch   5 | Train Loss: 0.339040 | Val Loss: 0.058967 | LR: 0.005000
Epoch   6 | Train Loss: 0.147994 | Val Loss: 0.053198 | LR: 0.003455
Epoch   7 | Train Loss: 0.146429 | Val Loss: 0.023571 | LR: 0.002061
Epoch   8 | Train Loss: 0.115384 | Val Loss: 0.033886 | LR: 0.000955
Epoch   9 | Train Loss: 0.100914 | Val Loss: 0.051012 | LR: 0.000245
Epoch  10 | Train Loss: 0.126045 | Val Loss: 0.076423 | LR: 0.000000

Final val_loss = 0.022136

==================================================
Pareto Front (Search Results)
==================================================
selection_method = knee_point
total_architectures_evaluated = 100

Top architectures:
  1. acc=0.6520, params=0.3000 [SELECTED]
  2. acc=0.6527, params=0.4291
  3. acc=0.6140, params=0.2934
