dof = 2*self.beta-1
return 1./np.sqrt(dof)*np.random.standard_t(dof, n)
if __name__ == '__main__':
if 0:
ntest = 100
pvals_calib = np.zeros(ntest)
pvals_direct = np.zeros(ntest)
for i in range(ntest):
data = np.random.randn(100)
pvals_calib[i] = calibrated_dip_test(data)
xF, yF = cum_distr(data)
dip = dip_from_cdf(xF, yF)
pvals_direct[i] = dip_pval_tabinterpol(dip, len(data))
from beeswarm import beeswarm
beeswarm([pvals_calib, pvals_direct]) # should be uniform distribution between 0 and 1
plt.show()
if 1:
import time
np.random.seed(1)
data = np.random.randn(100)
np.random.seed()
t0 = time.time()
pval = calibrated_dip_test(data)
t1 = time.time()
print("pval = {}".format(pval))
print("Time consumption calibrated dip test: {}".format(t1-t0))
fig, ax = plt.subplots()
for dp in dps:
query = """ SELECT Test_Item,Wafer_ID,Device,Unit,Measure FROM ACDC_RESULT
INNER JOIN TEST_FLOW ON ACDC_RESULT.ID = TEST_FLOW.ID
WHERE Desc='一致性' AND Test_Item =? AND Wafer_id=? AND DPS=?"""
dd = curs.execute(query, (tn, wd, dp))
ff = dd.fetchall()
# columns = ['测试项','wafer_id','芯片编号','单位','数据']
# data = pd.DataFrame(ff,columns = columns)
fdata = [i[4] for i in ff]
frame.append(fdata)
xticks.append("{}".format(dp))
colors.extend(GetColor(fdata))
Unit = ff[0][3]
bsw.beeswarm(frame, ax=ax, col=colors)
box = ax.boxplot(
frame,
positions=[0, 1, 2],
whis=1.5,
showfliers=False,
medianprops=medianprops,
meanprops=meanpointprops,
meanline=False,
showmeans=True,
)
plt.xticks(np.arange(3), xticks)
plt.xlabel("DPS (V)")
plt.ylabel("{} ({})".format(tn, Unit))
# plt.xlim(-2,5)
# plt.ylim(np.min(frame)-np.abs(np.max(frame)*0.1),np.max(frame)+np.abs(np.max(frame))*0.1)
def GetColor(x):
colors = []
for item in x:
if item > 0: colors.append("red")
else: colors.append("blue")
return colors
d1 = np.random.uniform(low=-3, high=3, size=100)
d2 = np.random.normal(size=100)
print(d1)
fig = plt.figure()
fig.set_size_inches((8,8))
ax1 = plt.subplot(221)
ax2 = plt.subplot(222)
ax3 = plt.subplot(223)
ax4 = plt.subplot(224)
colors = GetColor(d1) + GetColor(d2)
print(colors)
beeswarm([d1,d2], method="swarm", labels=["Uniform","Normal"], col=colors, ax=ax3)
beeswarm([d1,d2], method="swarm", labels=["Uniform","Normal"], col="black", ax=ax1)
print(d1)
beeswarm([d1,d2], method="swarm", labels=["Uniform","Normal"], col=["black","red"], ax=ax2)
#beeswarm([d1,d2], method="swarm", labels=["Uniform","Normal"], col=["red","blue","orange"], ax=ax4)
plt.tight_layout()
plt.show()
if __name__ == '__main__':
if 0:
ntest = 100
pvals_calib = np.zeros(ntest)
pvals_direct = np.zeros(ntest)
for i in range(ntest):
data = np.random.randn(100)
pvals_calib[i] = calibrated_dip_test(data)
xF, yF = cum_distr(data)
dip = dip_from_cdf(xF, yF)
pvals_direct[i] = dip_pval_tabinterpol(dip, len(data))
from beeswarm import beeswarm
beeswarm([pvals_calib, pvals_direct
]) # should be uniform distribution between 0 and 1
plt.show()
if 1:
import time
np.random.seed(1)
data = np.random.randn(100)
np.random.seed()
t0 = time.time()
pval = calibrated_dip_test(data)
t1 = time.time()
print("pval = {}".format(pval))
print("Time consumption calibrated dip test: {}".format(t1 - t0))
colors = []
q=np.percentile(x,[25,75])
w1 = q[1]+(q[1]-q[0])*1.5
w0 = q[0]-(q[1]-q[0])*1.5
x.sort()
for item in x:
if item > w1 or item < w0: colors.append("red")
else: colors.append("blue")
return colors
#
## Create data
np.random.seed(10)
collectn_1 = np.random.normal(100, 30, 200)
collectn_2 = np.random.normal(80, 30, 200)
collectn_3 = np.random.normal(90, 20, 200)
collectn_4 = np.random.normal(70, 25, 200)
## combine these different collections into a list
#data_to_plot = [collectn_1, collectn_2, collectn_3, collectn_4]
data_to_plot = [collectn_1]
# Create a figure instance
fig = plt.figure(1, figsize=(9, 6))
# Create an axes instance
ax = fig.add_subplot(111)
colors = GetColor(collectn_1)
# Create the boxplot
bp = ax.boxplot(data_to_plot,positions = [0])
bsw.beeswarm(data_to_plot,ax=ax,col=colors)
