A collection of our functions and classes from bootcamp.
Project description
JMI_MVM
- A collection of tools created for botcmap.
- More information to be added later.
Table of Contents
name = "JMI_MVM"
help_ = " Recommended Functions to try: \n calc_roc_auc & tune_params\n plot_hist_scat_sns & multiplot\n list2df & df_drop_regex\n plot_wide_kde_thin_bar & make_violinplot\n"
#functions.py
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl
import seaborn as sns
def calc_roc_auc(X_test,y_test,dtc,verbose=False):
"""Tests the results of an already-fit classifer.
Takes X_test, y_test, classifer, verbose (True" print result)
Returns the AUC for the roc_curve as a %"""
y_pred = dtc.predict(X_test)
FP_rate, TP_rate, thresh = roc_curve(y_test,y_pred)
roc_auc = auc(FP_rate,TP_rate)
roc_auc_perc = round(roc_auc*100,3)
# Your code here
if verbose:
print(f"roc_curve's auc = {roc_auc_perc}%")
return roc_auc_perc
def tune_params(param_name, param_values):
"""Takes in param_name to tune with param_values, plots train vs test AUC's.
Returns df_results and df_style with color coded results"""
res_list = [[param_name,'train_roc_auc','test_roc_auc']]
# Loop through all values in param_values
for value in param_values:
# Create Model, set params
dtc_temp = DecisionTreeClassifier(criterion='entropy')
params={param_name:value}
dtc_temp.set_params(**params)
# Fit model
dtc_temp.fit(X_train, y_train)
# Get roc_auc for training data
train_roc_auc = calc_roc_auc(X_train,y_train,dtc_temp)
# Get roc_auc for test data
test_res_roc_auc = calc_roc_auc(X_test,y_test,dtc_temp)
# Append value and results to res_list
res_list.append([value,train_roc_auc,test_res_roc_auc])
# Turn results into df_results (basically same as using list2df)
df_results = pd.DataFrame(res_list[1:],columns=res_list[0])
df_results.set_index(param_name,inplace=True)
# Plot df_results
df_results.plot()
# Color-coded dataframe s
import seaborn as sns
cm = sns.light_palette("green", as_cmap=True)
df_syle = df_results.style.background_gradient(cmap=cm)#,low=results.min(),high=results.max())
return df_results, df_syle
# MULTIPLOT
from string import ascii_letters
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
def multiplot(df):
"""Plots results from df.corr() in a correlation heat map for multicollinearity.
Returns fig, ax objects"""
sns.set(style="white")
# Compute the correlation matrix
corr = df.corr()
# Generate a mask for the upper triangle
mask = np.zeros_like(corr, dtype=np.bool)
mask[np.triu_indices_from(mask)] = True
# Set up the matplotlib figure
f, ax = plt.subplots(figsize=(16, 16))
# Generate a custom diverging colormap
cmap = sns.diverging_palette(220, 10, as_cmap=True)
# Draw the heatmap with the mask and correct aspect ratio
sns.heatmap(corr, mask=mask, annot=True, cmap=cmap, center=0,
square=True, linewidths=.5, cbar_kws={"shrink": .5})
return f, ax
# Plots histogram and scatter (vs price) side by side
# Plots histogram and scatter (vs price) side by side
def plot_hist_scat_sns(df, target='index'):
"""Plots seaborne distplots and regplots for columns im datamframe vs target.
Parameters:
df (DataFrame): DataFrame.describe() columns will be used.
target = name of column containing target variable.assume first coluumn.
Returns:
Figures for each column vs target with 2 subplots.
"""
import matplotlib.ticker as mtick
import matplotlib.pyplot as plt
import seaborn as sns
with plt.style.context(('dark_background')):
### DEFINE AESTHETIC CUSTOMIZATIONS -------------------------------##
# plt.style.use('dark_background')
figsize=(9,7)
# Axis Label fonts
fontTitle = {'fontsize': 14,
'fontweight': 'bold',
'fontfamily':'serif'}
fontAxis = {'fontsize': 12,
'fontweight': 'medium',
'fontfamily':'serif'}
fontTicks = {'fontsize': 8,
'fontweight':'medium',
'fontfamily':'serif'}
# Formatting dollar sign labels
fmtPrice = '${x:,.0f}'
tickPrice = mtick.StrMethodFormatter(fmtPrice)
### PLOTTING ----------------------------- ------------------------ ##
# Loop through dataframe to plot
for column in df.describe():
# print(f'\nCurrent column: {column}')
# Create figure with subplots for current column
fig, ax = plt.subplots(figsize=figsize, ncols=2, nrows=2)
## SUBPLOT 1 --------------------------------------------------##
i,j = 0,0
ax[i,j].set_title(column.capitalize(),fontdict=fontTitle)
# Define graphing keyword dictionaries for distplot (Subplot 1)
hist_kws = {"linewidth": 1, "alpha": 1, "color": 'blue','edgecolor':'w'}
kde_kws = {"color": "white", "linewidth": 1, "label": "KDE"}
# Plot distplot on ax[i,j] using hist_kws and kde_kws
sns.distplot(df[column], norm_hist=True, kde=True,
hist_kws = hist_kws, kde_kws = kde_kws,
label=column+' histogram', ax=ax[i,j])
# Set x axis label
ax[i,j].set_xlabel(column.title(),fontdict=fontAxis)
# Get x-ticks, rotate labels, and return
xticklab1 = ax[i,j].get_xticklabels(which = 'both')
ax[i,j].set_xticklabels(labels=xticklab1, fontdict=fontTicks, rotation=0)
ax[i,j].xaxis.set_major_formatter(mtick.ScalarFormatter())
# Set y-label
ax[i,j].set_ylabel('Density',fontdict=fontAxis)
yticklab1=ax[i,j].get_yticklabels(which='both')
ax[i,j].set_yticklabels(labels=yticklab1,fontdict=fontTicks)
ax[i,j].yaxis.set_major_formatter(mtick.ScalarFormatter())
# Set y-grid
ax[i, j].set_axisbelow(True)
ax[i, j].grid(axis='y',ls='--')
## SUBPLOT 2-------------------------------------------------- ##
i,j = 0,1
ax[i,j].set_title(column.capitalize(),fontdict=fontTitle)
# Define the kwd dictionaries for scatter and regression line (subplot 2)
line_kws={"color":"white","alpha":0.5,"lw":4,"ls":":"}
scatter_kws={'s': 2, 'alpha': 0.5,'marker':'.','color':'blue'}
# Plot regplot on ax[i,j] using line_kws and scatter_kws
sns.regplot(df[column], df[target],
line_kws = line_kws,
scatter_kws = scatter_kws,
ax=ax[i,j])
# Set x-axis label
ax[i,j].set_xlabel(column.title(),fontdict=fontAxis)
# Get x ticks, rotate labels, and return
xticklab2=ax[i,j].get_xticklabels(which='both')
ax[i,j].set_xticklabels(labels=xticklab2,fontdict=fontTicks, rotation=0)
ax[i,j].xaxis.set_major_formatter(mtick.ScalarFormatter())
# Set y-axis label
ax[i,j].set_ylabel(target,fontdict=fontAxis)
# Get, set, and format y-axis Price labels
yticklab = ax[i,j].get_yticklabels()
ax[i,j].set_yticklabels(yticklab,fontdict=fontTicks)
ax[i,j].yaxis.set_major_formatter(mtick.ScalarFormatter())
# ax[i,j].get_yaxis().set_major_formatter(tickPrice)
# Set y-grid
ax[i, j].set_axisbelow(True)
ax[i, j].grid(axis='y',ls='--')
## ---------- Final layout adjustments ----------- ##
# Deleted unused subplots
fig.delaxes(ax[1,1])
fig.delaxes(ax[1,0])
# Optimizing spatial layout
fig.tight_layout()
figtitle=column+'_dist_regr_plots.png'
# plt.savefig(figtitle)
return
# Tukey's method using IQR to eliminate
def detect_outliers(df, n, features):
"""Uses Tukey's method to return outer of interquartile ranges to return indices if outliers in a dataframe.
Parameters:
df (DataFrame): DataFrane containing columns of features
n: default is 0, multiple outlier cutoff
Returns:
Index of outliers for .loc
Examples:
Outliers_to_drop = detect_outliers(data,2,["col1","col2"]) Returning value
df.loc[Outliers_to_drop] # Show the outliers rows
data= data.drop(Outliers_to_drop, axis = 0).reset_index(drop=True)
"""
# Drop outliers
outlier_indices = []
# iterate over features(columns)
for col in features:
# 1st quartile (25%)
Q1 = np.percentile(df[col], 25)
# 3rd quartile (75%)
Q3 = np.percentile(df[col],75)
# Interquartile range (IQR)
IQR = Q3 - Q1
# outlier step
outlier_step = 1.5 * IQR
# Determine a list of indices of outliers for feature col
outlier_list_col = df[(df[col] < Q1 - outlier_step) | (df[col] > Q3 + outlier_step )].index
# append the found outlier indices for col to the list of outlier indices
outlier_indices.extend(outlier_list_col)
# select observations containing more than 2 outliers
outlier_indices = Counter(outlier_indices)
multiple_outliers = list( k for k, v in outlier_indices.items() if v > n )
return multiple_outliers
# describe_outliers -- calls detect_outliers
def describe_outliers(df):
""" Returns a new_df of outliers, and % outliers each col using detect_outliers.
"""
out_count = 0
new_df = pd.DataFrame(columns=['total_outliers', 'percent_total'])
for col in df.columns:
outies = detect_outliers(df[col])
out_count += len(outies)
new_df.loc[col] = [len(outies), round((len(outies)/len(df.index))*100, 2)]
new_df.loc['grand_total'] = [sum(new_df['total_outliers']), sum(new_df['percent_total'])]
return new_df
#### Cohen's d
def Cohen_d(group1, group2):
'''Compute Cohen's d.
# group1: Series or NumPy array
# group2: Series or NumPy array
# returns a floating point number
'''
diff = group1.mean() - group2.mean()
n1, n2 = len(group1), len(group2)
var1 = group1.var()
var2 = group2.var()
# Calculate the pooled threshold as shown earlier
pooled_var = (n1 * var1 + n2 * var2) / (n1 + n2)
# Calculate Cohen's d statistic
d = diff / np.sqrt(pooled_var)
return d
def plot_pdfs(cohen_d=2):
"""Plot PDFs for distributions that differ by some number of stds.
cohen_d: number of standard deviations between the means
"""
group1 = scipy.stats.norm(0, 1)
group2 = scipy.stats.norm(cohen_d, 1)
xs, ys = evaluate_PDF(group1)
pyplot.fill_between(xs, ys, label='Group1', color='#ff2289', alpha=0.7)
xs, ys = evaluate_PDF(group2)
pyplot.fill_between(xs, ys, label='Group2', color='#376cb0', alpha=0.7)
o, s = overlap_superiority(group1, group2)
print('overlap', o)
print('superiority', s)
def list2df(list):#, sort_values='index'):
""" Take in a list where row[0] = column_names and outputs a dataframe.
Keyword arguments:
set_index -- df.set_index(set_index)
sortby -- df.sorted()
"""
df_list = pd.DataFrame(list[1:],columns=list[0])
# df_list = df_list[1:]
return df_list
def df_drop_regex(DF, regex_list):
'''Use a list of regex to remove columns names. Returns new df.
Parameters:
DF -- input dataframe to remove columns from.
regex_list -- list of string patterns or regexp to remove.
Returns:
df_cut -- input df without the dropped columns.
'''
df_cut = DF.copy()
for r in regex_list:
df_cut = df_cut[df_cut.columns.drop(list(df_cut.filter(regex=r)))]
print(f'Removed {r}\n')
return df_cut
####### MIKE's PLOTTING
# plotting order totals per month in violin plots
def make_violinplot(x,y, title=None, hue=None, ticklabels=None):
'''Plots a violin plot with horizontal mean line, inner stick lines'''
plt.style.use('dark_background')
fig,ax =plt.subplots(figsize=(12,10))
sns.violinplot(x, y,cut=2,split=True, scale='count', scale_hue=True,
saturation=.5, alpha=.9,bw=.25, palette='Dark2',inner='stick', hue=hue).set_title(title)
ax.axhline(y.mean(),label='total mean', ls=':', alpha=.5, color='xkcd:yellow')
ax.set_xticklabels(ticklabels)
plt.legend()
plt.show()
x= df_year_orders['month']
y= df_year_orders['order_total']
title = 'Order totals per month with or without discounts'
hue=df_year_orders['Discount']>0
### Example usage
# #First, declare variables to be plotted
# x = df_year_orders['month']
# y = df_year_orders['order_total']
# ticks = [v for v in month_dict.values()]
# title = 'Order totals per month with or without discounts'
# hue = df_year_orders['Discount']>0
### Then call function
# make_violinplot(x,y,title,hue, ticks),
###########
def plot_wide_kde_thin_bar(series1,sname1, series2, sname2):
'''Plot series1 and series 2 on wide kde plot with small mean+sem bar plot.'''
## ADDING add_gridspec usage
import pandas as pd
import numpy as np
from scipy.stats import sem
import matplotlib.pyplot as plt
import matplotlib as mpl
import matplotlib.ticker as ticker
import seaborn as sns
from matplotlib import rcParams
from matplotlib import rc
rcParams['font.family'] = 'serif'
# Plot distributions of discounted vs full price groups
plt.style.use('default')
# with plt.style.context(('tableau-colorblind10')):
with plt.style.context(('seaborn-notebook')):
## ----------- DEFINE AESTHETIC CUSTOMIZATIONS ----------- ##
# Axis Label fonts
fontSuptitle ={'fontsize': 22,
'fontweight': 'bold',
'fontfamily':'serif'}
fontTitle = {'fontsize': 10,
'fontweight': 'medium',
'fontfamily':'serif'}
fontAxis = {'fontsize': 10,
'fontweight': 'medium',
'fontfamily':'serif'}
fontTicks = {'fontsize': 8,
'fontweight':'medium',
'fontfamily':'serif'}
## --------- CREATE FIG BASED ON GRIDSPEC --------- ##
plt.suptitle('Quantity of Units Sold', fontdict = fontSuptitle)
# Create fig object and declare figsize
fig = plt.figure(constrained_layout=True, figsize=(8,3))
# Define gridspec to create grid coordinates
gs = fig.add_gridspec(nrows=1,ncols=10)
# Assign grid space to ax with add_subplot
ax0 = fig.add_subplot(gs[0,0:7])
ax1 = fig.add_subplot(gs[0,7:10])
#Combine into 1 list
ax = [ax0,ax1]
### ------------------ SUBPLOT 1 ------------------ ###
## --------- Defining series1 and 2 for subplot 1------- ##
ax[0].set_title('Histogram + KDE',fontdict=fontTitle)
# Group 1: data, label, hist_kws and kde_kws
plotS1 = {'data': series1, 'label': sname1.title(),
'hist_kws' :
{'edgecolor': 'black', 'color':'darkgray','alpha': 0.8, 'lw':0.5},
'kde_kws':
{'color':'gray', 'linestyle': '--', 'linewidth':2,
'label':'kde'}}
# Group 2: data, label, hist_kws and kde_kws
plotS2 = {'data': series2,
'label': sname2.title(),
'hist_kws' :
{'edgecolor': 'black','color':'green','alpha':0.8 ,'lw':0.5},
'kde_kws':
{'color':'darkgreen','linestyle':':','linewidth':3,'label':'kde'}}
# plot group 1
sns.distplot(plotS1['data'], label=plotS1['label'],
hist_kws = plotS1['hist_kws'], kde_kws = plotS1['kde_kws'],
ax=ax[0])
# plot group 2
sns.distplot(plotS2['data'], label=plotS2['label'],
hist_kws=plotS2['hist_kws'], kde_kws = plotS2['kde_kws'],
ax=ax[0])
ax[0].set_xlabel(series1.name, fontdict=fontAxis)
ax[0].set_ylabel('Kernel Density Estimation',fontdict=fontAxis)
ax[0].tick_params(axis='both',labelsize=fontTicks['fontsize'])
ax[0].legend()
### ------------------ SUBPLOT 2 ------------------ ###
# Import scipy for error bars
from scipy.stats import sem
# Declare x y group labels(x) and bar heights(y)
x = [plotS1['label'], plotS2['label']]
y = [np.mean(plotS1['data']), np.mean(plotS2['data'])]
yerr = [sem(plotS1['data']), sem(plotS2['data'])]
err_kws = {'ecolor':'black','capsize':5,'capthick':1,'elinewidth':1}
# Create the bar plot
ax[1].bar(x,y,align='center', edgecolor='black', yerr=yerr,error_kw=err_kws,width=0.6)
# Customize subplot 2
ax[1].set_title('Average Quantities Sold',fontdict=fontTitle)
ax[1].set_ylabel('Mean +/- SEM ',fontdict=fontAxis)
ax[1].set_xlabel('')
ax[1].tick_params(axis=y,labelsize=fontTicks['fontsize'])
ax[1].tick_params(axis=x,labelsize=fontTicks['fontsize'])
ax1=ax[1]
test = ax1.get_xticklabels()
labels = [x.get_text() for x in test]
ax1.set_xticklabels([plotS1['label'],plotS2['label']], rotation=45,ha='center')
# xlab = [x.get_text() for x in xlablist]
# ax[1].set_xticklabels(xlab,rotation=45)
# fig.savefig('H1_EDA_using_gridspec.png')
# plt.tight_layout()
# print(f')
plt.show()
return fig,ax
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
JMI_MVM-0.3.4.tar.gz
(41.5 kB
view details)
Built Distribution
JMI_MVM-0.3.4-py3-none-any.whl
(65.6 kB
view details)
File details
Details for the file JMI_MVM-0.3.4.tar.gz
.
File metadata
- Download URL: JMI_MVM-0.3.4.tar.gz
- Upload date:
- Size: 41.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5e819eb0e64a7294e90f27b8c9d21e142ca11b3202402a1016155cec37c3f4dc |
|
MD5 | add6261816be7d1396c08729e51554ca |
|
BLAKE2b-256 | 2c8193d0dcde8641c7785f13d50de7ad1117ebf74856ad41f2790e6b31e80883 |
File details
Details for the file JMI_MVM-0.3.4-py3-none-any.whl
.
File metadata
- Download URL: JMI_MVM-0.3.4-py3-none-any.whl
- Upload date:
- Size: 65.6 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/1.13.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.0.0 requests-toolbelt/0.9.1 tqdm/4.31.1 CPython/3.7.2
File hashes
Algorithm | Hash digest | |
---|---|---|
SHA256 | 9393deadb243a2bb736f6b76e5ccc9d4ee0f666b4881e5434a4c5dd9e80eea6e |
|
MD5 | e7b9e1a78b3f1e744867e66f8646e072 |
|
BLAKE2b-256 | dfe6e365eb6d6862394911cd8f7b7234a2e80ebf64bb0ce54bf37d5a71bee7d6 |