"global_batch_size": det.Constant(value=32),
"dense1": det.Constant(value=128),
},
"searcher": {"name": "single", "metric": "val_accuracy", "max_steps": 40},
}
config.update(json.loads(args.config))
context = init(config, local=args.local, test=args.test, context_dir=str(pathlib.Path.cwd()))
train_images, train_labels = data.load_training_data()
train_images = train_images / 255.0
train_data = _ArrayLikeAdapter(
x=train_images, y=train_labels, batch_size=context.get_per_slot_batch_size()
)
test_images, test_labels = data.load_validation_data()
test_images = test_images / 255.0
test_data = _ArrayLikeAdapter(
x=test_images, y=test_labels, batch_size=context.get_per_slot_batch_size()
)
model = build_model(context)
if args.use_fit:
model.fit(
x=train_images,
y=train_labels,
batch_size=context.get_per_slot_batch_size(),
validation_data=test_data,
use_multiprocessing=False,
workers=1,
def build_validation_data_loader(self) -> InputData:
test_images, test_labels = data.load_validation_data()
test_images = test_images / 255.0
return test_images, test_labels
import numpy as np
import pickle
from evaluation import load_model, process_data, run
from data import load_validation_data, pair_superset
# Load Model and Training Data
model = load_model()
(x_val, y_val) = load_validation_data()
# Save Predictions
predictions = []
# create test positive and negative pairs
te_pairs, te_y = pair_superset(x_val, y_val)
for i in range(len(te_pairs[:1000])):
print(f'Running Test {i}')
data = process_data(te_pairs[i][0], te_pairs[i][1])
# print(f'data Shape : {data.shape}')
result = run(data, model)
predictions.append(result)