Transforms
In [1]:
Copied!
# transforms
# transforms
In [2]:
Copied!
from scipy.stats import lognorm, rv_histogram, norm
import numpy as np
import seaborn as sns
from volumetricspy.stats import to_normal, NScaler
from scipy.stats import lognorm, rv_histogram, norm
import numpy as np
import seaborn as sns
from volumetricspy.stats import to_normal, NScaler
In [3]:
Copied!
data = lognorm.rvs(s = 0.8, scale=300, size=1000)
sns.displot(data, kde=False, rug=True)
data = lognorm.rvs(s = 0.8, scale=300, size=1000)
sns.displot(data, kde=False, rug=True)
Out[3]:
<seaborn.axisgrid.FacetGrid at 0x7f7bc031fe20>
In [4]:
Copied!
ts = to_normal(data, loc=0, scale=1)
sns.displot(ts, kde=False, rug=True)
ts = to_normal(data, loc=0, scale=1)
sns.displot(ts, kde=False, rug=True)
Out[4]:
<seaborn.axisgrid.FacetGrid at 0x7f7bc09185e0>
In [5]:
Copied!
ts.mean()
ts.mean()
Out[5]:
-0.00040551182342109726
In [6]:
Copied!
ts.std()
ts.std()
Out[6]:
1.0293784093796814
In [7]:
Copied!
type(np.histogram(ts, bins=100))
type(np.histogram(ts, bins=100))
Out[7]:
tuple
In [8]:
Copied!
nsc = NScaler(bins=100).fit(data)
tsc = nsc.transform(data)
sns.displot(tsc)
nsc = NScaler(bins=100).fit(data)
tsc = nsc.transform(data)
sns.displot(tsc)
Out[8]:
<seaborn.axisgrid.FacetGrid at 0x7f7bc031fdf0>
In [9]:
Copied!
sns.displot(nsc.inverse(tsc))
sns.displot(nsc.inverse(tsc))
Out[9]:
<seaborn.axisgrid.FacetGrid at 0x7f7bf42a3460>
In [12]:
Copied!
data.shape
data.shape
Out[12]:
(1000,)
In [13]:
Copied!
nsc.inverse(tsc).shape
nsc.inverse(tsc).shape
Out[13]:
(1000,)
In [ ]:
Copied!