num_classes=num_classes,
num_dense_layers=num_dense_layers,
num_dense_units=num_dense_units,
pooling=pooling,
dropout_rate=dropout_rate,
kernel_regularizer=dense_layer_regularizer,
save_to=run_name,
load_from=prev_run_name,
print_model_summary=True,
plot_model_summary=False,
lr=lr)
n_samples_train = x_train.shape[0]
n_samples_valid = x_valid.shape[0]
class_weights = compute_class_weights(y_train, wt_type=class_wt_type)
batch_size = 32
use_data_aug = True
horizontal_flip = True
vertical_flip = True
rotation_angle = 180
width_shift_range = 0.1
height_shift_range = 0.1
log_variable(var_name='num_dense_layers', var_value=num_dense_layers)
log_variable(var_name='num_dense_units', var_value=num_dense_units)
log_variable(var_name='dropout_rate', var_value=dropout_rate)
log_variable(var_name='pooling', var_value=pooling)
log_variable(var_name='class_wt_type', var_value=class_wt_type)
log_variable(var_name='dense_layer_regularizer',
print("x_train shape: ", x_train.shape)
print("y_train shape: ", y_train.shape)
print("x_valid shape: ", x_valid.shape)
print("y_valid shape: ", y_valid.shape)
"""
(x_train, y_train), (x_valid, y_valid), _ = load_training_data(task_idx=3,
output_size=224,
num_partitions=1,
test_split = 0.8)"""
class_wt_type = 'balanced'
run_name = 'OriginalResNet152'
num_classes = y_train.shape[1]
class_weights = compute_class_weights(y, wt_type=class_wt_type)
print(class_weights)
callbacks = config_cls_callbacks(run_name)
n_samples_train = x_train.shape[0]
n_samples_valid = x_valid.shape[0]
sys.stdout.flush()
"""#Size of training batch
batch_train = x_train.shape[0]
#Size of testing batch
batch_test = x_test.shape[0]
y_train = y_train.flatten()
