Skip to content
This repository was archived by the owner on Jan 2, 2021. It is now read-only.

make mean and stdev class-level variables in test class #26

Open
bfgray3 wants to merge 1 commit into
base: master
Choose a base branch
from
Open

make mean and stdev class-level variables in test class #26

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
68 changes: 23 additions & 45 deletions tests/test_bootstrap.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25,8 +25,12 @@ def to_sparse(data):
mat[0, :len(data)] = data
return mat


class BootstrappedTest(unittest.TestCase):

mean = 100
stdev = 10

def setUp(self):
np.random.seed(1)
import warnings
Expand All @@ -38,18 +42,15 @@ def test_bootstrap_results(self):
self.assertEqual(bsr.error_fraction(), np.inf)
self.assertEqual((bsr + 1).error_fraction(), 2)

self.assertEqual(bsr.is_significant(), False)
self.assertEqual((bsr + 2).is_significant(), True)
self.assertEqual((bsr - 2).is_significant(), True)
self.assertFalse(bsr.is_significant())
self.assertTrue((bsr + 2).is_significant())
self.assertTrue((bsr - 2).is_significant())
self.assertEqual(bsr.get_result(), 0)
self.assertEqual((bsr + 2).get_result(), 1)
self.assertEqual((bsr - 2).get_result(), -1)

def test_result_math(self):
mean = 100
stdev = 10

samples = np.random.normal(loc=mean, scale=stdev, size=5000)
samples = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)

bsr = bs.bootstrap(samples, bs_stats.mean)

Expand All @@ -63,10 +64,7 @@ def test_result_math(self):
self.assertEqual(2 * bsr.value, (2 * bsr).value)

def test_bootstrap(self):
mean = 100
stdev = 10

samples = np.random.normal(loc=mean, scale=stdev, size=5000)
samples = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)

bsr = bs.bootstrap(samples, bs_stats.mean)

Expand All @@ -81,11 +79,9 @@ def test_bootstrap(self):
self.assertTrue(bsr.lower_bound < bsr2.lower_bound)

def test_bootstrap_ab(self):
mean = 100
stdev = 10

test = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl = np.random.normal(loc=mean, scale=stdev, size=5000)
test = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test = test * 1.1

bsr = bs.bootstrap_ab(test, ctrl, bs_stats.mean,
Expand All @@ -112,8 +108,8 @@ def test_bootstrap_ab(self):
delta=.5
)

test_denom = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl_denom = np.random.normal(loc=mean, scale=stdev, size=5000)
test_denom = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl_denom = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test_denom = test_denom * 1.1

bsr4 = bs.bootstrap_ab(test, ctrl, bs_stats.mean,
Expand Down Expand Up @@ -234,11 +230,8 @@ def test_randomized_permutation_ratio(self):
)

def test_bootstrap_batch_size(self):
mean = 100
stdev = 10

test = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl = np.random.normal(loc=mean, scale=stdev, size=5000)
test = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test = test * 1.1

bsr = bs.bootstrap_ab(test, ctrl, bs_stats.mean,
Expand Down Expand Up @@ -288,11 +281,8 @@ def test_bootstrap_batch_size(self):
)

def test_bootstrap_threads(self):
mean = 100
stdev = 10

test = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl = np.random.normal(loc=mean, scale=stdev, size=5000)
test = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test = test * 1.1

bsr = bs.bootstrap_ab(test, ctrl, bs_stats.mean,
Expand Down Expand Up @@ -342,11 +332,8 @@ def test_bootstrap_threads(self):
)

def test_pivotal(self):
mean = 100
stdev = 10

test = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl = np.random.normal(loc=mean, scale=stdev, size=5000)
test = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test = test * 1.1

bsr = bs.bootstrap_ab(test, ctrl, bs_stats.mean,
Expand Down Expand Up @@ -396,10 +383,7 @@ def test_pivotal(self):
)

def test_bootstrap_sparse(self):
mean = 100
stdev = 10

samples = np.random.normal(loc=mean, scale=stdev, size=5000)
samples = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
samples_sp = sparse.csr_matrix(samples)

bsr = bs.bootstrap(samples, bs_stats.mean)
Expand Down Expand Up @@ -430,11 +414,8 @@ def test_bootstrap_sparse(self):
self.assertAlmostEqual(bsr.lower_bound, bsr_sp.lower_bound, delta=.2)

def test_bootstrap_ab_sparse(self):
mean = 100
stdev = 10

test = np.random.normal(loc=mean, scale=stdev, size=500)
ctrl = np.random.normal(loc=mean, scale=stdev, size=5000)
test = np.random.normal(loc=self.mean, scale=self.stdev, size=500)
ctrl = np.random.normal(loc=self.mean, scale=self.stdev, size=5000)
test = test * 1.1
test_sp = sparse.csr_matrix(test)
ctrl_sp = sparse.csr_matrix(ctrl)
Expand Down Expand Up @@ -462,13 +443,10 @@ def test_bootstrap_ab_sparse(self):
)

def test_t_dist(self):
mean = 100
stdev = 100

sample_size = [250, 500, 1000, 2500, 3500, 5000, 8000, 10000]

for i in sample_size:
samples = np.random.normal(loc=mean, scale=stdev, size=i)
samples = np.random.normal(loc=self.mean, scale=self.stdev, size=i)
bsr = bs.bootstrap(samples, stat_func=bs_stats.mean, alpha=0.05)

mr = st.t.interval(1 - 0.05, len(samples) - 1, loc=np.mean(samples),
Expand Down