Younes-Charfaoui

val intent = Intent(Intent.ACTION_SEND)
val activityList = packageManager.queryIntentActivities(intent, PackageManager.MATCH_ALL)
when {
    activityList.size > 1 -> {
        val chooser = Intent.createChooser(intent, "Choose an App")
        startActivity(chooser)
    }
    intent.resolveActivity(packageManager) != null -> startActivity(intent)
    else -> Toast.makeText(this, "No App to launch with", Toast.LENGTH_LONG).show()
}

Younes-Charfaoui

<?xml version="1.0" encoding="utf-8"?>
<network-security-config>
    <!-- Don't allow clear text traffic for all domains unless is not in the domain-config -->
    <base-config cleartextTrafficPermitted="false" />

    <!-- Special configuration to some domains -->
    <domain-config cleartextTrafficPermitted="true">
        <domain includeSubdomains="true">localhost</domain>
        <trust-anchors>
            <!-- Trust a debug certificate in addition to the system certificates -->

Younes-Charfaoui

import numpy as np

def haversine_distance(row):
    lat_p, lon_p = row['Pickup Lat'], row['Pickup Long']
    lat_d, lon_d = row['Destination Lat'], row['Destination Long']
    radius = 6371 # km

    dlat = np.radians(lat_d - lat_p)
    dlon = np.radians(lon_d - lon_p)
    a = np.sin(dlat/2) * np.sin(dlat/2) + np.cos(np.radians(lat_p)) * np.cos(np.radians(lat_d)) * np.sin(dlon/2) * np.sin(dlon/2)

Younes-Charfaoui

from sklearn.cluster import KMeans ,AgglomerativeClustering

# creates 5 clusters using hierarchical clustering.
agc = AgglomerativeClustering(n_clusters =5, affinity='euclidean', linkage='ward')
train['pickup cluster'] = agc.fit_predict(train[['Pickup Lat','Pickup Long']])

# creates 5 clusters using k-means clustering algorithm.
kmeans = KMeans(5)
clusters = kmeans.fit_predict(train[['Pickup Lat','Pickup Long']])
train['pickup cluster'] = kmeans.predict(train[['Pickup Lat','Pickup Long']])

Younes-Charfaoui

import folium

# create the map.
map_pickup = folium.Map( location=[-1.317755,36.830370])

# adding the latitude and longitude points to the map.
train.apply(lambda row:folium.CircleMarker(location=[row["Pickup Lat"], row["Pickup Long"]] ).add_to(map_pickup), axis=1)

# display the map: just ask for the object representation in juypter notebook.
map_pickup

Younes-Charfaoui

from geopy.geocoders import Nominatim

# create the locator
geolocator = Nominatim(user_agent="myGeocoder")

# getting the location address
location = geolocator.reverse("52.509669, 13.376294")
print(location)
# >>> result : Backwerk, Potsdamer Platz, Tiergarten, Mitte, Berlin, 10785, Deutschland

Younes-Charfaoui

import numpy as np

train['pickup x'] = np.cos(train['Pickup Lat']) * np.cos(train['Pickup Long'])
train['pickup y'] = np.cos(train['Pickup Lat']) * np.sin(train['Pickup Long'])
train['pickup z'] = np.sin(train['Pickup Lat'])

Younes-Charfaoui

# import and create the VarianceThreshold object.
from sklearn.feature_selection import VarianceThreshold
vs_constant = VarianceThreshold(threshold=0)

# select the numerical columns only.
numerical_x_train = x_train[x_train.select_dtypes([np.number]).columns]

# fit the object to our data.
vs_constant.fit(numerical_x_train)

Younes-Charfaoui

# transpose the feature matrice
train_features_T = x_train.T

# print the number of duplicated features
print(train_features_T.duplicated().sum())

# select the duplicated features columns names
duplicated_columns = train_features_T[train_features_T.duplicated()].index.values

# drop those columns

Younes-Charfaoui

# make a threshold for quasi constant.
threshold = 0.98

# create empty list
quasi_constant_feature = []

# loop over all the columns
for feature in x_train.columns:

    # calculate the ratio.