plt.figure(figsize=(12,6))
sns.histplot(df['SalePrice'])

plt.figure(figsize=(15,8))
sns.lineplot(data=df[:100]['SalePrice'])

plt.figure(figsize=(12,6))
sns.violinplot(x = df['SaleCondition'], y = df['SalePrice'])
plt.axhline(df[df['SaleCondition'] == 'Normal']['SalePrice'].mean(),\
            color='r',linestyle='dashed',label='normal_avg')
plt.legend()

plt.figure(figsize=(12,6))
sns.boxplot(data=df, y='SaleCondition', x='SalePrice', orient='h')

plt.figure(figsize=(12,6))
sns.boxplot(data=df, x='SaleCondition', y='SalePrice', orient='v')

ridge_plot = sns.FacetGrid(df, row="SaleCondition", hue="SaleCondition", aspect=5, height=1.25)  
ridge_plot.map(sns.kdeplot, 'SalePrice', shade=True)
ridge_plot.map(plt.axhline)
ridge_plot.fig.subplots_adjust(hspace=0.35)

from seaborn_qqplot import pplot

pplot(df.iloc[:250,:], x='YearBuilt', y='SalePrice', kind='qq', height=4, aspect=2)

sns.set_style('darkgrid')

g = sns.catplot(x="sex", y="count",
                hue="survived", col="pclass",
                data=df, kind="bar",
                height=6, aspect=.7, palette="flare");

df_copy2  = df['sex'].value_counts().reset_index()
df_copy2.columns = ['gender', 'count']
df = df_copy2

plt.figure(figsize=(8,4))
plt.title('Titanic Gender Distribution')

sns.barplot(x='gender', y='count', data=df, palette='pastel', alpha=0.9)

plt.title('Titanic Gender Distribution')
plt.bar(x=df['gender'], height=df['count'], color=['blue', 'red'], alpha=0.4, width=0.4)
plt.xlabel('Gender')
plt.ylabel('Count')

# Flip the x and y variables
plt.figure(figsize=(8,4))
plt.title('Titanic Gender Distribution')
sns.barplot(x='count', y='gender', data=df, palette='pastel', alpha=0.5)

# y and width are the passed params
plt.title('Titanic Gender Distribution')
plt.barh(y=df['gender'], width=df['count'], color=['blue', 'red'], alpha=0.4)
plt.xlabel('Gender')
plt.ylabel('Count')

# notice the order parameter
plt.figure(figsize=(8,4))
plt.title('Titanic Gender Distribution')

sns.barplot(x='gender', y='count', data=df, palette='pastel', alpha=0.9, order=['male', 'female'])

plt.figure(figsize=(20,10))
plt.hlines(y=my_range, xmin=0, xmax=df['fare'], color='skyblue')
plt.grid(True)
plt.plot(df['fare'], my_range, "o")
plt.yticks(my_range, df['name'])
plt.title("Ticket Price Dot Plot", loc='left')
plt.xlabel('Ticket Price')
plt.ylabel('Name')

plt.figure(figsize=(20,10))
plt.hlines(y=my_range, xmin=0, xmax=df['fare'], color='skyblue')
plt.hlines(y=my_range, xmin=0, xmax=df['age'], color='red')

plt.grid(True)
plt.plot(df['fare'], my_range, "o")
plt.plot(df['age'], my_range, "o")

plt.yticks(my_range, ordered_df['name'])
plt.title("Ticket Price Dot Plot", loc='left')
plt.xlabel('Ticket Price')
plt.ylabel('Name')

plt.figure(figsize=(12,6), dpi=140)
num_col1 = 'BMI'
num_col2 = 'BloodPressure'
target= 'Outcome'
cat_num_col1='Pregnancies'
cat_num_col2 ='Age'

sns.scatterplot(x=num_col1, y=num_col2, data=df,
                style=target, hue=cat_num_col1, 
                size=cat_num_col2, alpha=0.7, palette = 'plasma',
)#,sizes=(20,100), hue_norm=(0,15))

plt.figure(figsize=(12,6), dpi=140)
num_col1 = 'BMI'
num_col2 = 'BloodPressure'
target= 'Outcome'
cat_num_col1='Pregnancies'
cat_num_col2 ='Age'

sns.lmplot(x=num_col1, y=num_col2, markers=['o','x'], hue=target, data=df, fit_reg=True)

num_col1 = 'BMI'
num_col2 = 'BloodPressure'
target= 'Outcome'
cat_num_col1='Pregnancies'
cat_num_col2 ='Age'

sns.relplot(
    data=df, x=num_col1, y=num_col2,
    col=target, hue=cat_num_col1, size=cat_num_col2, style = target,palette = 'plasma',
    kind="scatter"#,aspect=0.5, height=12
)

plt.figure(dpi=140)
sns.pairplot(df)
plt.show()

plt.figure(dpi = 140)
sns.pairplot(df,hue = 'Outcome',palette = 'plasma')
plt.legend(['Non Diabetic','Diabetic'])
plt.show()

sns.kdeplot(data=df, x=num_col1, y=num_col2, hue=target,fill=True,alpha=0.5,palette = 'plasma')

plt.figure(figsize= (14,8))
# cmap=sns.diverging_palette(5, 250, as_cmap=True)
cmap = sns.diverging_palette(250, 10, as_cmap=True)
ax = sns.heatmap(df.corr(),center = 0,annot= True,linewidth=0.5,cmap= cmap)

corr = df.corr()
plt.figure(figsize=(14,8))
cmap=sns.diverging_palette(5, 250, as_cmap=True)
mask = np.zeros_like(corr)
mask[np.triu_indices_from(mask)] = True
with sns.axes_style("white"):
    ax = sns.heatmap(corr, mask=mask, vmax=.3, square=True,cmap=cmap,center = 0,annot=True)

plt.figure(dpi = 80,figsize= (14,8))
mask = np.triu(np.ones_like(df.corr(),dtype = bool))
sns.heatmap(df.corr(),mask = mask, fmt = ".2f",annot=True,lw=1,cmap = 'plasma')
plt.yticks(rotation = 0)
plt.xticks(rotation = 90)
plt.title('Correlation Heatmap')
plt.show()

plt.figure(dpi = 100, figsize = (5,4))
comparsion_variable = 'Glucose'
target = 'Outcome'

print("Joint plot of {} with Other Variables ==> \n".format(comparsion_variable))
for i in  df.columns:
    if i != comparsion_variable and i != target:
        print(f"Correlation between {comparsion_variable} and {i} ==> ",df.corr().loc[comparsion_variable][i])
        sns.jointplot(x=comparsion_variable,y=i,data=df,kind = 'reg',color = 'purple')
        plt.show()

from pandas.plotting import parallel_coordinates

numeric_cols = ['BloodPressure','Pregnancies','BMI','SkinThickness','Glucose',target]

tdf = df.sample(100)

parallel_coordinates(tdf[numeric_cols], target, color = ['r','b'])

fig, axs = plt.subplots(ncols=1, sharey=True, figsize=(10, 6))
fig.subplots_adjust(hspace=0.5, left=0.07, right=0.93)
ax = axs
hb = ax.hexbin(df["age"], df["fare"], gridsize=5, cmap='Blues', alpha = 0.9)
ax.axis([min(df['age']), max(df['age']), min(df['fare']), max(df['fare'])])
ax.set_title("Hexagon binning")
cb = fig.colorbar(hb, ax=ax)
cb.set_label('counts')
ax.set_xlabel('Age')
ax.set_ylabel('Fare')

x = sns.jointplot(data=df, x="age", y="fare", kind="hist")
x.ax_joint.set_title("Square Binplots with distributions", pad=70.0)
cb = fig.colorbar(hb, ax=x.ax_marg_y)
cb.set_label('counts')
print(ax)
plt.show()

x = sns.jointplot(data=df, x="age", y="fare", kind="hex")
x.ax_marg_y.axis([min(df['age']), max(df['age']), min(df['fare']), max(df['fare'])])
cb = fig.colorbar(hb, ax=x.ax_marg_y)
cb.set_label('counts')

from statsmodels.graphics.mosaicplot import mosaic

#df = pd.DataFrame({'size' : ['small', 'large', 'large', 'small', 'large', 'small', 'large', 'large'], 'length' : ['long', 'short', 'medium', 'medium', 'medium', 'short', 'long', 'medium'], 'temp' : ['cold', 'hot', 'cold', 'warm', 'warm', 'cold', 'hot', 'warm']})

props = {}
single_low = 28
max_start = 255
max_start_oth = 229
diff = 25
r,g,b=max_start,max_start_oth,max_start_oth
for x in df['sex'].unique(): #unique colums in each
    for y in df['pclass'].unique():
        col = '#{}{}{}'.format(format(int(r),'02x'),format(int(g),'02x'),format(int(b),'02x'))
        for z in df['survived'].unique():
            props[(str(z), str(y), str(x))] ={'color': col}
            if r==max_start:
                g-=diff
                b-=diff
            elif b==max_start:
                r-=diff
                g-=diff
            elif g==max_start:
                r-=diff
                b-=diff
            if (g<single_low and b<single_low):
                b,r,g=max_start,max_start_oth,max_start_oth
            elif (r<single_low and g<single_low):
                g,b,r = max_start,max_start_oth,max_start_oth
            elif (r<single_low and b<single_low):
                print("no more colors")

import matplotlib as mpl
from statsmodels.graphics.mosaicplot import mosaic
mpl.rc("figure", figsize=(14,5))
mosaic(df, ['survived', 'pclass', 'sex'], properties=props, title='Survival of Passengers on Titanic - Mosaic ')
plt.show()

plt.figure(figsize= (14,8))
# cmap = cmap=sns.diverging_palette(5, 250, as_cmap=True)
cmap = sns.diverging_palette(250, 10, as_cmap=True)
ax = sns.heatmap(df.corr(),center = 0,annot= True,linewidth=0.5,cmap= cmap)
plt.title('Heatmap for categorical variables in Titanic Dataset', size=15)

corr = df.corr()
plt.figure(figsize=(14,8))
mask = np.zeros_like(corr)
mask[np.triu_indices_from(mask)] = True
with sns.axes_style("white"):
    ax = sns.heatmap(corr, mask=mask, vmax=.3, square=True,cmap=cmap,center = 0)
plt.title('Heatmap for categorical variables in Titanic Dataset - Partial Matrix', size=15)

survived_columns = ['pclass', 'survived', 'sibsp', 'parch']
for col in survived_columns:
    val = df[col].value_counts(dropna=False)
    if(len(val.index)>100):
        print("Too many Categories in "+col)
        continue
    sns.barplot(x=val.index,y=val.values,label=True, alpha=0.8)
    plt.title(col)
    plt.ylabel('Count')
    plt.xlabel("Classes")
    plt.grid('True')
    plt.show()

import itertools
default = df[df["survived"]==1]
non_default = df[df["survived"]==0]

d_cols =['pclass', 'survived', 'sibsp', 'parch']
d_length = len(d_cols)

fig = plt.figure(figsize=(16,4))
for i,j in itertools.zip_longest(d_cols,range(d_length)):
    plt.subplot(1,4,j+1)
    default[i].value_counts().plot.pie(autopct = "%1.0f%%",colors = sns.color_palette("prism"),startangle = 90,
                                        wedgeprops={"linewidth":1,"edgecolor":"white"},shadow =True)
    circ = plt.Circle((0,0),.7,color="white")
    plt.gca().add_artist(circ)
    plt.ylabel("")
    plt.title(i+"-Survivor")


fig = plt.figure(figsize=(16,4))
for i,j in itertools.zip_longest(d_cols,range(d_length)):
    plt.subplot(1,4,j+1)
    non_default[i].value_counts().plot.pie(autopct = "%1.0f%%",colors = sns.color_palette("prism",3),startangle = 90,
                                           wedgeprops={"linewidth":1,"edgecolor":"white"},shadow =True)
    circ = plt.Circle((0,0),.7,color="white")
    plt.gca().add_artist(circ)
    plt.ylabel("")
    plt.title(i+"-Dead")

categorical_columns = ['pclass', 'survived', 'sibsp', 'parch']
target = "survived"

for col in categorical_columns:
	plt.figure(figsize=(16,8))
	plt.subplot(121)
	df[df[target]==0][col].value_counts().plot.pie(fontsize=9,autopct = "%1.0f%%",colors = sns.color_palette("Set1"),
	wedgeprops={"linewidth":2,"edgecolor":"white"},shadow =True)
	circ = plt.Circle((0,0),.7,color="white")
	plt.gca().add_artist(circ)
	plt.title("Distribution of "+col+" type for target==0",color="b")

	plt.subplot(122)
	df[df[target]==1][col].value_counts().plot.pie(fontsize=9,autopct = "%1.0f%%", colors = sns.color_palette("Set1"),
	wedgeprops={"linewidth":2,"edgecolor":"white"},shadow =True)
	circ = plt.Circle((0,0),.7,color="white")
	plt.gca().add_artist(circ)
	plt.title("Distribution of "+col+" type for target==1",color="b")
	plt.ylabel("")
	plt.show()

#!conda install -c conda-forge geopy --yes 
from geopy.geocoders import Nominatim # convert an address into latitude and longitude values

#!conda install -c conda-forge folium=0.5.0 --yes
import folium # map rendering library

address = 'New York, USA'

geolocator = Nominatim(user_agent="nyc_explorer")
location = geolocator.geocode(address)
latitude = location.latitude
longitude = location.longitude

map_nyc = folium.Map(location=[latitude, longitude], zoom_start=12)

map_nyc

cu_lat = 40.8075
cu_lng = -73.9626

map_cu = folium.Map(location=[cu_lat, cu_lng], zoom_start=16)

folium.CircleMarker(
    [cu_lat, cu_lng],
    radius=10,
    color='blue',
    fill=True,
    fill_color='#3186cc',
    fill_opacity=0.7,
    parse_html=False).add_to(map_cu)

map_cu

#!conda install -c plotly plotly
import plotly.express as px
import plotly.offline as py
py.init_notebook_mode(connected=True)

df_px = px.data.election()
geojson = px.data.election_geojson()

fig = px.choropleth(df_px, geojson=geojson, color="Bergeron",
                    locations="district", featureidkey="properties.district",
                    projection="mercator"
                   )
fig.update_geos(fitbounds="locations", visible=False)
fig.update_layout(margin={"r":0,"t":0,"l":0,"b":0})
#fig.show() # Use this to render the plot in your notebook

py.plot({"data": fig}, output_type="div", show_link="False", include_plotlyjs="False", link_text="") # For HTML rendering

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline

import plotly.offline as py
import plotly.graph_objects as go
py.init_notebook_mode(connected=True)

df = pd.read_csv("https://raw.githubusercontent.com/plotly/datasets/master/school_earnings.csv")

data = [go.Bar(x=df.School,
            y=df.Gap)]
layout = go.Layout(dict(title = "Male and Female gap in schools",
                  xaxis = dict(title = 'Number of students'),
                  yaxis = dict(title = 'School Name'),
                  ))

py.plot(dict(data=data, layout=layout), include_plotlyjs=False, output_type='div')

# py.iplot(dict(data=data, layout=layout),filename='Basic bar plot')

%matplotlib inline
sns.set_style('darkgrid')

f, ax = plt.subplots(nrows=6, ncols=1, figsize=(15, 30))

for i, col in enumerate(df.drop('date', axis=1).columns):
    sns.lineplot(x='date', y=col,data=df, ax=ax[i], color='dodgerblue')
    ax[i].set_title('Feature: {}'.format(col), fontsize=14)
    ax[i].set_ylabel(ylabel=col, fontsize=14)

plt.figure(figsize=(12,6))
sns.lineplot(data=df[['adj_close','open','date']].set_index('date'))

# A year has 52 weeks (52 weeks * 7 days per week) aporx.
rolling_window = 52
f = plt.figure(figsize=(15, 6))
ax = plt.gca()

sns.lineplot(x=df['date'], y=df['adj_close'], color='dodgerblue')
sns.lineplot(x=df['date'], y=df['adj_close'].rolling(rolling_window).mean(),  color='black', label='rolling mean')
sns.lineplot(x=df['date'], y=df['adj_close'].rolling(rolling_window).std(), color='orange', label='rolling std')
ax.set_title('Adjusted Close: Non-stationary \nnon-constant mean & non-constant variance', fontsize=14)
ax.set_ylabel(ylabel='Adjusted Close Price', fontsize=14)
ax.set_xlim([pd.to_datetime('2020-01-01', format='%Y-%m-%d'), pd.to_datetime('2020-12-31', format='%Y-%m-%d')])

plt.tight_layout()
plt.show()

from statsmodels.tsa.seasonal import seasonal_decompose

core_columns =  ['adj_close','volume']

for column in core_columns:
    decomp = seasonal_decompose(df[column], period=52, model='additive', extrapolate_trend='freq')
    df[f"{column}_trend"] = decomp.trend
    df[f"{column}_seasonal"] = decomp.seasonal

fig, ax = plt.subplots(ncols=2, nrows=4, sharex=True, figsize=(16,8))

for i, column in enumerate(['adj_close', 'volume']):
    
    res = seasonal_decompose(df[column], freq=52, model='additive', extrapolate_trend='freq')

    ax[0,i].set_title('Decomposition of {}'.format(column), fontsize=16)
    res.observed.plot(ax=ax[0,i], legend=False, color='dodgerblue')
    ax[0,i].set_ylabel('Observed', fontsize=14)

    res.trend.plot(ax=ax[1,i], legend=False, color='dodgerblue')
    ax[1,i].set_ylabel('Trend', fontsize=14)

    res.seasonal.plot(ax=ax[2,i], legend=False, color='dodgerblue')
    ax[2,i].set_ylabel('Seasonal', fontsize=14)
    
    res.resid.plot(ax=ax[3,i], legend=False, color='dodgerblue')
    ax[3,i].set_ylabel('Residual', fontsize=14)

plt.show()

f, ax = plt.subplots(nrows=2, ncols=1, figsize=(15, 12))
f.suptitle('Seasonal Components of Features', fontsize=16)

for i, column in enumerate(core_columns):
    sns.lineplot(x=df['date'], y=df[column + '_seasonal'], ax=ax[i], color='dodgerblue', label='P25')
    ax[i].set_ylabel(ylabel=column, fontsize=14)
    ax[i].set_xlim([pd.to_datetime('2020-01-01', format='%Y-%m-%d'), pd.to_datetime('2020-12-31', format='%Y-%m-%d')])
    
plt.tight_layout()
plt.show()