def test_fit_one_shot_model_epoch(self):
rs = RandomSearcher(self.get_space_simple,
optimize_direction=OptimizeDirection.Maximize)
hk = HyperKeras(rs,
optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'],
callbacks=[SummaryCallback()],
one_shot_mode=True,
one_shot_train_sampler=rs)
x, y = self.get_x_y_1()
hk.fit_one_shot_model_epoch(x, y)
def test_build_dataset_iter(self):
rs = RandomSearcher(self.get_space,
optimize_direction=OptimizeDirection.Maximize)
hk = HyperKeras(rs,
optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'],
callbacks=[SummaryCallback()])
x, y = self.get_x_y_1()
ds_iter = hk.build_dataset_iter(x, y, batch_size=10)
batch_counter = 0
for x_b, y_b in ds_iter:
# x_b, y_b = next()
assert len(x_b) == 10
assert len(y_b) == 10
batch_counter += 1
assert batch_counter == 10
ds_iter = hk.build_dataset_iter(x, y, batch_size=32)
batch_counter = 0
for x_b, y_b in ds_iter:
# x_b, y_b = next()
if batch_counter < 3:
assert len(x_b) == 32
assert len(y_b) == 32
else:
assert len(x_b) == 4
assert len(y_b) == 4
batch_counter += 1
assert batch_counter == 4
ds_iter = hk.build_dataset_iter(x, y, batch_size=32, repeat_count=2)
batch_counter = 0
for x_b, y_b in ds_iter:
# x_b, y_b = next()
if batch_counter < 6:
assert len(x_b) == 32
assert len(y_b) == 32
else:
assert len(x_b) == 8
assert len(y_b) == 8
batch_counter += 1
assert batch_counter == 7
def test_model_with_hp(self):
rs = RandomSearcher(self.get_space,
optimize_direction=OptimizeDirection.Maximize)
hk = HyperKeras(rs,
optimizer='adam',
loss='sparse_categorical_crossentropy',
metrics=['accuracy'],
callbacks=[SummaryCallback()])
x, y = self.get_x_y()
hk.search(x, y, x, y, max_trails=3)
best_trial = hk.get_best_trail()
estimator = hk.final_train(best_trial.space_sample, x, y)
score = estimator.predict(x)
result = estimator.evaluate(x, y)
assert len(score) == 100
assert result
# model2 = space.keras_model(deepcopy=False)
# model2.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
# model2.fit(x_train[:samples], y_train[:samples], batch_size=32)
# result2 = model.evaluate(x_train[:samples], y_train[:samples])
#
# weights_cache = LayerWeightsCache()
# space = enas_micro_search_space(arch='NR', hp_dict={}, use_input_placeholder=False, weights_cache=weights_cache)
# space.random_sample()
#
# model = SharingWeightModel(space)
# model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
# model.fit(x_train[:samples], y_train[:samples], batch_size=32)
# result = model.evaluate(x_train[:samples], y_train[:samples])
#
# space = enas_micro_search_space(arch='NR', hp_dict={}, use_input_placeholder=False, weights_cache=weights_cache)
# space.random_sample()
# model.update_search_space(space)
# model.fit(x_train[:samples], y_train[:samples], batch_size=100)
# result = model.evaluate(x_train[:samples], y_train[:samples])
rs = RandomSearcher(
lambda: enas_micro_search_space(arch='NNRNNR', hp_dict={}),
optimize_direction='max')
hk = HyperKeras(rs, optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'],
callbacks=[SummaryCallback()], one_shot_mode=True, visualization=False)
# tenserboard = TensorBoard('./tensorboard/run_enas')
hk.search(x_train[:samples], y_train[:samples], x_test[:int(samples / 10)], y_test[:int(samples / 10)],
max_trails=100, epochs=1, callbacks=[])
assert hk.get