jaradc

import numpy as np
from scipy.stats import entropy
from math import log, e
import pandas as pd

import timeit

def entropy1(labels, base=None):
  value,counts = np.unique(labels, return_counts=True)
  return entropy(counts, base=base)

jaradc

def specificity_scorer(y_true, y_pred):
  tn, fp, fn, tp = metrics.confusion_matrix(y_true, y_pred).ravel()
  return tn / (tn+fp)

specificity = metrics.make_scorer(specificity_scorer, greater_is_better=True)

# use in cross_val_score
cv = cross_val_score(xgb, dt.features_.values, y, scoring=specificity, cv=5, verbose=2)

# use in GridSearchCV

jaradc

import statsmodels.api as sm
import numpy as np
import rpy2.robjects.packages as rpackages
import rpy2.robjects as robjects
rstats = rpackages.importr('stats')

s1 = 1556
n1 = 2455

s2 = 1671

jaradc

>>> corr
               Impressions    Clicks  CTR  Avg. CPC      Cost  Avg. position  \
Impressions       1.000000  0.599646  NaN  0.301556  0.568137       0.197353   
Clicks            0.599646  1.000000  NaN  0.566357  0.987073       0.627268   
CTR                    NaN       NaN  NaN       NaN       NaN            NaN   
Avg. CPC          0.301556  0.566357  NaN  1.000000  0.663789       0.809944   
Cost              0.568137  0.987073  NaN  0.663789  1.000000       0.707918   
Avg. position     0.197353  0.627268  NaN  0.809944  0.707918       1.000000   
Conversions       0.558450  0.927165  NaN  0.717237  0.962034       0.746493   

jaradc

import pandas as pd
import statsmodels.api as sm
from statsmodels.tsa.stattools import adfuller

def test_stationarity(timeseries):
  '''
  timeseries has a date index and a single column of numeric values
  '''
    
    #Determing rolling statistics

jaradc

df = pd.DataFrame(
	{'Label': [np.nan, np.nan, 'Label1', 'Label2'],
	 'URL': ['https://www.website.com/where-to-buy',
		 np.nan, np.nan, 'https://www.website.com/store']
	 }, columns=['Label', 'URL'])

# this does not actually replace inplace!
df[['Label', 'URL']].fillna('', inplace=True)

# you have to assign it or use a dict

jaradc

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.colors import ListedColormap

# A random colormap for matplotlib
rand_cmap = ListedColormap(np.random.rand(256,3))

for i in range(5):
  plt.scatter([1,2,3], np.random.randn(3), s=10, cmap=rand_cmap)
  

jaradc

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from scipy.optimize import curve_fit

df = pd.DataFrame({
    'y': [0.996559203, 0.99161362, 0.9925214090000001, 0.986498352,
                0.9826329420000001, 0.977550635, 0.9542758440000001, 0.941359915,
                0.933388103, 0.929990698, 0.920058004, 0.90789857, 0.909764261,
                0.8944469829999999, 0.912682288, 0.913135466, 0.913485262,

jaradc

def ma_subplots(df, window, title=None):
	fig, ax = plt.subplots(df.shape[1], 1, figsize=(12, 7))
	ax = ax.ravel()
	for i,col in enumerate(df):
		mean = df[col].mean()
		ma = df[col].rolling(window).mean()
		mstd = df[col].rolling(window).std()
		ax[i].plot(df.index, df[col], color='k', label=col)
		ax[i].plot(ma.index, ma, 'b', label='Moving Avg.')
		ax[i].fill_between(mstd.index, ma - 2*mstd, ma + 2*mstd, color='b', alpha=0.2)

jaradc

def scale_between(series, min_amt, max_amt):
	series_min = series.min()
	series_max = series.max()
	return (((max_amt - min_amt)*(series - series_min)) / (series_max - series_min)) + min_amt