add continent graph

data.py

@@ -0,0 +1,54 @@
+countries = {'fr': 'france',
+             'us': 'united states',
+             'usa': 'united states',
+             'uk': 'united kingdom',
+             'ge': 'germany'}
+
+country_ref = {
+    'france': 'europe',
+    'ireland': 'europe',
+    'luxembourg': 'europe',
+    'japan': 'asia',
+    'germany': 'europe',
+    'united states': 'north america',
+    'netherlands': 'europe',
+    'italy': 'europe', 'switzerland': 'europe',
+    'brazil': 'south america',
+    'sweden': 'europe',
+    'venezuela': 'south america', 'spain': 'europe', 'norway': 'europe', 'finland': 'europe',
+    'panama': 'south america',
+    'united kingdom': 'europe',
+    'denmark': 'europe',
+    'australia': 'australia',
+    'cayman islands': 'north america',
+    'saudi arabia': 'asia',
+    'hungary': 'europe',
+    'belgium': 'europe',
+    'iran (islamic republic of)': 'asia',
+    'canada': 'north america',
+    'morocco': 'africa',
+    'luxemburg': 'europe',
+    'liechtenstein': 'europe',
+    'curacao': 'south america',
+    'russia': 'europe',
+    'hong-kong': 'asia',
+    'albania': 'europe',
+    'chile': 'south america',
+    'china': 'asia',
+    'south korea': 'asia',
+    'israel': 'asia',
+    'taiwan': 'asia',
+    'singapore': 'asia',
+    'austria': 'europe',
+    'china (mainland)': 'asia',
+    'hong kong': 'asia',
+    'british virgin islands': 'europe',
+    'dubai': 'asia',
+    'ecuador': 'south america',
+    'korea': 'asia',
+    'mexico': 'south america',
+    'saoudi arabia': 'asia',
+    'indonesia': 'asia',
+    'united arab emirates': 'asia',
+    'seychelles': 'africa',
+    'costa rica': 'south america'}

data_processing.py

@@ -1,6 +1,9 @@
 import streamlit as st
 import requests
 import pandas as pd
+import numpy as np
+
+from data import countries, country_ref
 
 
 # util functions
@@ -52,22 +55,31 @@ def process_data(data):
     # work on organisations
     organizations = pd.DataFrame(data['organizations'])
     organizations['id'] = organizations.url.apply(get_id)
+    organizations.country = organizations.country.str.lower().str.strip().apply(lambda v: countries.get(v, v))
     organizations = organizations[["id", "name", "company_type", "revenues", "currency", "country", "lei"]]
+    organizations['continent'] = organizations.country.apply(lambda v: country_ref.get(v,v))
     organizations.columns = ['org_' + col for col in organizations.columns]
     decisions = decisions.merge(organizations, left_on='organizations', right_on='org_id')
-
     # remove Individual
     decisions = decisions[decisions.org_company_type != "Individual"]
-    decisions.org_country = decisions.org_country.str.lower().str.strip()
 
     # work on authorities
     authorities = pd.DataFrame(data['authorities'])
     authorities.index = authorities.url.apply(get_id)
     authorities = authorities[["country", "type", "name"]]
-    countries = {'FR': 'france', 'US': 'United States', 'UK': 'United Kingdom', 'GE': 'Germany'}
-    authorities.country = authorities.country.apply(lambda v: countries.get(v, v)).str.lower().str.strip()
+    authorities.country = authorities.country.str.lower().str.strip().apply(lambda v: countries.get(v, v))
 
     decisions['authorities_name'] = decisions.authorities.apply(replace_lis_val(authorities, 'name'))
     decisions['authorities_country'] = decisions.authorities.apply(replace_lis_val(authorities, 'country')).apply(mode)
 
     return decisions, organizations, authorities
+
+
+def get_monetary_dataframe(decision_scope):
+    monetary_decision = decision_scope[decision_scope.monetary_sanction > 0]
+    monetary_decision = monetary_decision[monetary_decision.org_revenues != ""]
+    monetary_decision['org_revenues'] = monetary_decision.org_revenues.astype(float)
+    monetary_decision['log10_org_revenues'] = monetary_decision.org_revenues.apply(np.log10)
+    monetary_decision['log10_monetary_sanction'] = monetary_decision.monetary_sanction.apply(np.log10)
+    monetary_decision['same_country'] = (monetary_decision.org_country == monetary_decision.authorities_country)
+    return monetary_decision

stream_app.py

@@ -1,10 +1,9 @@
 # -*- coding: utf-8 -*-
 import streamlit as st
 
-import numpy as np
 import plotly.express as px
 
-from data_processing import load_data, process_data
+from data_processing import load_data, process_data, get_monetary_dataframe
 
 st.title("Data Analysis 🌎 📃")
 st.write("by [Teolex](https://www.theolex.io/)")
@@ -27,29 +26,47 @@ if authority != 'All':
     authority_filter = decisions.authorities_name.apply(lambda a: authority in a)
 else:
     authority_filter = decisions.authorities_name.apply(lambda a: bool(set(select_auth) & set(a)))
+year_filter = (decisions.year >= min_year) & (decisions.year <= max_year)
+decision_scope = decisions[authority_filter & year_filter]
 
+# explore monetary sanctions
+monetary_decision = get_monetary_dataframe(decision_scope)
 
-decisions.head()
-year_filter = (decisions.year >= min_year) & (decisions.year <= max_year)
+##
+# Plot Graphs
+##
+
+st.subheader("The organizations' sectors targeted by the sanctions: ")
+st.markdown("The graph shows the cumulated monetary sanction for the current filters")
+
+fig = px.treemap(monetary_decision,
+                 path=['org_company_type'],
+                 color='org_revenues',
+                 color_continuous_scale='RdBu',
+                 template="simple_white",
+                 values='monetary_sanction')
+st.plotly_chart(fig)
+
+st.subheader("The organizations' regions targeted by the sanctions: ")
+st.markdown("The graph shows the cumulated monetary sanction for the current filters")
+fig = px.treemap(monetary_decision[~monetary_decision.org_continent.isnull()],
+                 path=['org_continent','org_country'],
+                 color='org_revenues',
+                 color_continuous_scale='RdBu',
+                 template="simple_white",
+                 values='monetary_sanction')
+st.plotly_chart(fig)
 
-decision_scope = decisions[authority_filter & year_filter]
 
-## explore monetary sanctions
-monetary_decision = decision_scope[decision_scope.monetary_sanction > 0]
-monetary_decision = monetary_decision[monetary_decision.org_revenues != ""]
-monetary_decision['org_revenues'] = monetary_decision.org_revenues.astype(float)
-monetary_decision['log10_org_revenues'] = monetary_decision.org_revenues.apply(np.log10)
-monetary_decision['log10_monetary_sanction'] = monetary_decision.monetary_sanction.apply(np.log10)
-monetary_decision['same_country'] = (monetary_decision.org_country == monetary_decision.authorities_country)
+st.subheader("Revenues vs monetary sanctions representation ")
+st.markdown("The graph shows the cumulated monetary sanction for the current filters")
 fig = px.scatter(monetary_decision,
                  x="org_revenues",
                  y="monetary_sanction",
                  log_x=True,
-                 log_y=True,template="simple_white",
-                 color="authorities_country",
-                 trendline="ols",
+                 log_y=True,
+                 template="simple_white",
+                 color="same_country",
+                 color_continuous_scale='RdBu',
                  hover_name="org_name")
 st.plotly_chart(fig)
-st.markdown("Comments...")
-st.subheader("1.1 Main takeaways: ")
-st.write("sample text.")