def test_measure_raise(self):
ctx = [
dict(x_name2=i,
va=numpy.arange(i),
vb=numpy.arange(10) - 5,
pydot=pydot) for i in range(10, 10000, 100)
]
with self.assertRaises(ValueError):
list(measure_time_dim('pydot(va, vb)', ctx, verbose=1))
def test_measure_time_dim(self):
ctx = [
dict(x_name=i,
va=numpy.arange(i),
vb=numpy.arange(10) - 5,
pydot=pydot) for i in range(10, 10000, 100)
]
ms = list(measure_time_dim('pydot(va, vb)', ctx, verbose=1))
self.assertIsInstance(ms, list)
self.assertEqual(len(ms), 100)
ctxs = [
dict(va=numpy.random.randn(n).astype(dtype),
vb=numpy.random.randn(n).astype(dtype),
dot=fct,
x_name=n) for n in range(10, n, h)
]
return ctxs
##############################
# numpy dot
# +++++++++
#
ctxs = get_vectors(numpy.dot, 10000)
df = DataFrame(list(measure_time_dim('dot(va, vb)', ctxs, verbose=1)))
df['fct'] = 'numpy.dot'
print(df.tail(n=3))
dfs = [df]
##############################
# Several cython dot
# ++++++++++++++++++
#
for fct in [
dot_product, ddot_cython_array, ddot_cython_array_optim, ddot_array,
ddot_array_16, ddot_array_16_sse
]:
ctxs = get_vectors(fct, 10000 if fct.__name__ != 'dot_product' else 1000)
dfs = []
sets = list(range(2, 145, 20))
##############################
# numpy mul
# +++++++++
#
ctxs = [
dict(va=numpy.random.randn(n, n).astype(numpy.float64),
vb=numpy.random.randn(n, n).astype(numpy.float64),
mul=lambda x, y: x @ y,
x_name=n) for n in sets
]
res = list(measure_time_dim('mul(va, vb)', ctxs, verbose=1))
dfs.append(DataFrame(res))
dfs[-1]['fct'] = 'numpy'
pprint.pprint(dfs[-1].tail(n=2))
##############################
# Simple multiplication
# +++++++++++++++++++++
#
ctxs = [
dict(va=numpy.random.randn(n, n).astype(numpy.float64),
vb=numpy.random.randn(n, n).astype(numpy.float64),
mul=dmul_cython_omp,
x_name=n) for n in sets
]
##############################
# python dot: pydot
# +++++++++++++++++
#
# The first function :func:`pydot
# <td3a_cpp.tutorial.pydot>` uses
# python to implement the dot product.
ctxs = [
dict(va=numpy.random.randn(n).astype(numpy.float64),
vb=numpy.random.randn(n).astype(numpy.float64),
pydot=pydot,
x_name=n) for n in range(10, 1000, 100)
]
res_pydot = list(measure_time_dim('pydot(va, vb)', ctxs, verbose=1))
pprint.pprint(res_pydot[:2])
##############################
# numpy dot
# +++++++++
#
ctxs = [
dict(va=numpy.random.randn(n).astype(numpy.float64),
vb=numpy.random.randn(n).astype(numpy.float64),
dot=numpy.dot,
x_name=n) for n in range(10, 50000, 100)
]
def numpy_filter(va, mx):
va[va > mx] = mx
all_res = []
for fct in [numpy_filter,
pyfilter_dmax, filter_dmax_cython,
filter_dmax_cython_optim,
cyfilter_dmax,
cfilter_dmax, cfilter_dmax2,
cfilter_dmax16, cfilter_dmax4]:
print(fct)
ctxs = get_vectors(fct, 1000 if fct == pyfilter_dmax else 40000)
res = list(measure_time_dim('fil(va, mx)', ctxs, verbose=1))
for r in res:
r['fct'] = fct.__name__
all_res.extend(res)
pprint.pprint(all_res[:2])
#############################
# Let's display the results
# +++++++++++++++++++++++++
cc = DataFrame(all_res)
cc['N'] = cc['x_name']
fig, ax = plt.subplots(2, 2, figsize=(10, 10))
cc[cc.N <= 1100].pivot('N', 'fct', 'average').plot(
y = y.astype(numpy.float64)
fct = ElasticNet(alpha=alpha, l1_ratio=L1_ratio)
ctxs = [
dict(X=X, y=y, linear_regression=fct.fit, x_name=n)
for n in range(10, n, h)
]
return ctxs
## Get time execution
# sklearn ElasticNet
ctxs = get_vectors_elastic(100000)
df = DataFrame(
list(measure_time_dim('linear_regression(X, y)', ctxs, verbose=1)))
df['fct'] = 'ElasticNet'
print(df.tail(n=3))
dfs = [df]
# naive & c++ implemantation
for fct in [nv_regular_linreg, cpp_regular_linreg]:
ctxs = get_vectors(fct, 100000)
df = DataFrame(
list(
measure_time_dim(
'linear_regression(X, y, beta, alpha, L1_ratio, max_iter, tol, num_samples, num_features)',
ctxs,
verbose=1)))
def get_vectors_sklearn(n, h=10, dtype=numpy.float64):
X, y = make_blobs(n_samples=n, centers=3, n_features=4)
_, p = X.shape
X = X.astype(numpy.float64)
y = y.astype(numpy.float64)
fct = RandomForestClassifier(n_estimators=5, max_depth=2, max_features=2)
ctxs = [dict(X=X,
y=y,
RandomForestClassifier=fct.fit,
x_name=n
)
for n in range(10, n, h)]
return ctxs
ctxs = get_vectors_sklearn(100)
df = DataFrame(list(measure_time_dim('RandomForestClassifier(X,y)', ctxs, verbose=1)))
df['fct'] = 'RandomForestClassifier'
print(df.tail(n=3))
dfs = [df]
ctxs = get_vectors_cpp(create_forest_prime, 100)
df = DataFrame(list(measure_time_dim('random_forest(data, n_trees, max_depth, max_x)', ctxs, verbose=1)))
df['fct'] = create_forest_prime.__name__
dfs.append(df)
print(df.tail(n=3))
ctxs = get_vectors_nv(random_forest, 100)
df = DataFrame(list(measure_time_dim('random_forest(data, max_depth, min_size, n_trees, max_x)', ctxs, verbose=1)))
df['fct'] = random_forest.__name__
dfs.append(df)