def get_next_memory_windows(trace_file, cache_type, cache_size, n_early_stop,
seq_start, parameters, args_global):
# use 80% warmup, manually check for unimodal behavior
n_warmup = int(0.8 * n_early_stop)
# per cache size parameters overwrite other parameters
df = database.load_reports(
trace_file=trace_file,
cache_type=cache_type,
cache_size=str(cache_size),
seq_start=str(seq_start),
n_early_stop=str(n_early_stop),
n_warmup=n_warmup,
version=parameters['version'], # use version as a strong key
dburi=parameters["dburi"],
dbcollection=parameters["dbcollection"],
)
if len(df) == 0:
next_windows = np.linspace(1,
int(0.4 * n_early_stop),
args_global['n_beam'] + 1,
dtype=int)[1:]
else:
# window at most 40% of length
parameter_space = ParameterSpace(
[ContinuousParameter('x', 1, int(0.4 * n_early_stop))])
bo = GPBayesianOptimization(
variables_list=parameter_space.parameters,
X=df['memory_window'].values.reshape(-1, 1),
Y=df['byte_miss_ratio'].values.reshape(-1, 1),
batch_size=args_global['n_beam'])
next_windows = bo.suggest_new_locations().reshape(-1).astype(int)
return next_windows
def get_validation_tasks_per_cache_size(trace_file, cache_type, cache_size, parameters, args_global, df):
n_req = parameters['n_req']
n_validation = int(n_req * args_global['ratio_validation'])
n_iteration = args_global['n_iteration']
n_beam = args_global['n_beam']
if len(df) == 0 or len(df[df['cache_size'] == cache_size]) == 0:
# init value
next_windows = np.linspace(1, int(0.4 * n_validation), args_global['n_beam'] + 1, dtype=int)[1:]
tasks = []
for memory_window in next_windows:
# override n_early stop
task = _get_task(trace_file, cache_type, cache_size, parameters, n_validation, memory_window)
tasks.append(task)
return tasks
# as emukit output is random, don't assume same points each time, as long as # point is enough
df1 = df[df['cache_size'] == cache_size]
if len(df1) >= n_iteration * n_beam:
return []
# next round
# window at most 40% of length
# add xs, ys in a consistent order otherwise the results will be difference
parameter_space = ParameterSpace([ContinuousParameter('x', 1, int(0.4 * n_validation))])
bo = GPBayesianOptimization(variables_list=parameter_space.parameters,
X=df1['memory_window'].values.reshape(-1, 1),
Y=df1['byte_miss_ratio'].values.reshape(-1, 1),
batch_size=args_global['n_beam'])
next_windows = bo.suggest_new_locations().reshape(-1).astype(int)
tasks = []
for memory_window in next_windows:
task = _get_task(trace_file, cache_type, cache_size, parameters, n_validation, memory_window)
tasks.append(task)
return tasks
def test_loop():
n_iterations = 5
x_init = np.random.rand(5, 1)
y_init = np.random.rand(5, 1)
x = ContinuousParameter('x', 0, 1)
bo = GPBayesianOptimization(variables_list=[x], X=x_init, Y=y_init)
bo.run_optimization(f, n_iterations)
# Check we got the correct number of points
assert bo.loop_state.X.shape[0] == n_iterations + 5