def lasso():
"""Fit Lasso."""
print("Fitting LAS...", end=" ", flush=True)
time.sleep(SLEEP)
t0 = time.time()
ls = Lasso()
ls.fit(X, y)
print_time(t0, "Done", end="")
def knn():
"""Fit KNN."""
print("Fitting KNN...", end=" ", flush=True)
time.sleep(SLEEP)
t0 = time.time()
knn = KNeighborsRegressor()
knn.fit(X, y)
print_time(t0, "Done", end="")
def ensemble():
"""Fit ensemble."""
print("Fitting ENS...", end=" ", flush=True)
time.sleep(SLEEP)
t0 = time.time()
ens = build_ensemble(shuffle=False, folds=2)
ens.fit(X, y)
print_time(t0, "Done", end="")
def elasticnet():
"""Fit Elastic Net."""
print("Fitting ELN...", end=" ", flush=True)
time.sleep(SLEEP)
t0 = time.time()
ls = Lasso()
ls.fit(X, y)
print_time(t0, "Done", end="")
name = e.__class__.__name__
e = clone(e)
t0 = perf_counter()
e.fit(X, y)
t1 = perf_counter() - t0
times[n][name].append(t1)
print('%s (%i) : %6.2fs |' % (name, n, t1),
end=" ",
flush=True)
print()
print()
print_time(ts, "Benchmark done")
if PLOT:
try:
import matplotlib.pyplot as plt
plt.ion()
print("Plotting results...", end=" ", flush=True)
plt.figure(figsize=(8, 8))
x = range(STEP, MAX + STEP, STEP)
cm = [
plt.cm.rainbow(i)
for i in np.linspace(0, 1.0, int(3 * len(cores)))
]
@profile
def elasticnet():
"""Fit Elastic Net."""
print("Fitting ELN...", end=" ", flush=True)
time.sleep(SLEEP)
t0 = time.time()
ls = Lasso()
ls.fit(X, y)
print_time(t0, "Done", end="")
if __name__ == '__main__':
X, y = make_friedman1(MAX, COLS)
print("\nML-ENSEMBLE\n")
print("Benchmark of ML-ENSEMBLE memory profile against "
"Scikit-learn estimators.\n"
"Data shape: (%i, %i)\n"
"Data size: %i MB\n" % (MAX, COLS, np.ceil(X.nbytes / 1e+6)))
ts = time.time()
lasso()
knn()
ensemble()
elasticnet()
print_time(ts, "\nProfiling complete.")