Jan Mühlnikel
commited on
Commit
·
e3302f1
1
Parent(s):
55a6bd8
added country and orga filter
Browse files
__pycache__/similarity_page.cpython-310.pyc
CHANGED
|
Binary files a/__pycache__/similarity_page.cpython-310.pyc and b/__pycache__/similarity_page.cpython-310.pyc differ
|
|
|
functions/__pycache__/filter_projects.cpython-310.pyc
CHANGED
|
Binary files a/functions/__pycache__/filter_projects.cpython-310.pyc and b/functions/__pycache__/filter_projects.cpython-310.pyc differ
|
|
|
functions/filter_projects.py
CHANGED
|
@@ -4,8 +4,11 @@ def contains_code(crs_codes, code_list):
|
|
| 4 |
codes = str(crs_codes).split(';')
|
| 5 |
return any(code in code_list for code in codes)
|
| 6 |
|
| 7 |
-
def filter_projects(df, crs3_list, crs5_list, sdg_str):
|
|
|
|
| 8 |
if crs3_list != [] or crs5_list != [] or sdg_str != "":
|
|
|
|
|
|
|
| 9 |
if crs3_list and not crs5_list:
|
| 10 |
df = df[df['crs_3_code'].apply(lambda x: contains_code(x, crs3_list))]
|
| 11 |
elif crs3_list and crs5_list:
|
|
@@ -13,9 +16,24 @@ def filter_projects(df, crs3_list, crs5_list, sdg_str):
|
|
| 13 |
elif not crs3_list and crs5_list:
|
| 14 |
df = df[df['crs_5_code'].apply(lambda x: contains_code(x, crs5_list))]
|
| 15 |
|
|
|
|
| 16 |
if sdg_str != "":
|
| 17 |
df = df[df["sgd_pred_code"] == int(sdg_str)]
|
| 18 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
return df
|
| 20 |
|
| 21 |
|
|
|
|
| 4 |
codes = str(crs_codes).split(';')
|
| 5 |
return any(code in code_list for code in codes)
|
| 6 |
|
| 7 |
+
def filter_projects(df, crs3_list, crs5_list, sdg_str, country_code_list, orga_code_list):
|
| 8 |
+
# Check if filters where not all should be selected are empty
|
| 9 |
if crs3_list != [] or crs5_list != [] or sdg_str != "":
|
| 10 |
+
|
| 11 |
+
# FILTER CRS
|
| 12 |
if crs3_list and not crs5_list:
|
| 13 |
df = df[df['crs_3_code'].apply(lambda x: contains_code(x, crs3_list))]
|
| 14 |
elif crs3_list and crs5_list:
|
|
|
|
| 16 |
elif not crs3_list and crs5_list:
|
| 17 |
df = df[df['crs_5_code'].apply(lambda x: contains_code(x, crs5_list))]
|
| 18 |
|
| 19 |
+
# FILTER SDG
|
| 20 |
if sdg_str != "":
|
| 21 |
df = df[df["sgd_pred_code"] == int(sdg_str)]
|
| 22 |
|
| 23 |
+
# FILTER COUNTRY
|
| 24 |
+
if country_code_list != []:
|
| 25 |
+
country_filtered_df = pd.DataFrame()
|
| 26 |
+
for c in country_code_list:
|
| 27 |
+
c_df = df[df["country"].str.contains(c, na=False)]
|
| 28 |
+
country_filtered_df = pd.concat([country_filtered_df, c_df], ignore_index=True)
|
| 29 |
+
|
| 30 |
+
df = country_filtered_df
|
| 31 |
+
|
| 32 |
+
# FILTER ORGANIZATION
|
| 33 |
+
if orga_code_list != []:
|
| 34 |
+
df = df[df['orga_abbreviation'].isin(orga_code_list)]
|
| 35 |
+
|
| 36 |
+
|
| 37 |
return df
|
| 38 |
|
| 39 |
|
modules/filter_modules.py
DELETED
|
@@ -1,21 +0,0 @@
|
|
| 1 |
-
import pandas as pd
|
| 2 |
-
import streamlit as st
|
| 3 |
-
|
| 4 |
-
def country_option(special_cases, country_names):
|
| 5 |
-
country_option = st.multiselect(
|
| 6 |
-
'Country / Countries',
|
| 7 |
-
special_cases + country_names,
|
| 8 |
-
placeholder="Select"
|
| 9 |
-
)
|
| 10 |
-
|
| 11 |
-
return country_option
|
| 12 |
-
|
| 13 |
-
def orga_option(special_cases, orga_names):
|
| 14 |
-
orga_list = special_cases + [f"{v[0]} ({k})" for k, v in orga_names.items()]
|
| 15 |
-
orga_option = st.multiselect(
|
| 16 |
-
'Development Bank / Organization',
|
| 17 |
-
orga_list,
|
| 18 |
-
placeholder="Select"
|
| 19 |
-
)
|
| 20 |
-
|
| 21 |
-
return orga_option
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
similarity_page.py
CHANGED
|
@@ -79,6 +79,18 @@ def getSDG():
|
|
| 79 |
|
| 80 |
return SDG_NAMES
|
| 81 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
# Load Sentence Transformer Model
|
| 83 |
@st.cache_resource
|
| 84 |
def load_model():
|
|
@@ -110,6 +122,8 @@ CRS3_MERGED = getCRS3()
|
|
| 110 |
CRS5_MERGED = getCRS5()
|
| 111 |
SDG_NAMES = getSDG()
|
| 112 |
|
|
|
|
|
|
|
| 113 |
model = load_model()
|
| 114 |
sentences, embeddings, faiss_index = load_embeddings_and_index()
|
| 115 |
|
|
@@ -153,7 +167,25 @@ def show_page():
|
|
| 153 |
|
| 154 |
|
| 155 |
with col2:
|
| 156 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 157 |
|
| 158 |
|
| 159 |
# CRS CODE LIST
|
|
@@ -166,8 +198,14 @@ def show_page():
|
|
| 166 |
else:
|
| 167 |
sdg_str = ""
|
| 168 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 169 |
# FILTER DF WITH SELECTED FILTER OPTIONS
|
| 170 |
-
filtered_df = filter_projects(projects_df, crs3_list, crs5_list, sdg_str)
|
| 171 |
|
| 172 |
# FIND MATCHES
|
| 173 |
p1_df, p2_df = calc_matches(filtered_df, projects_df, sim_matrix)
|
|
|
|
| 79 |
|
| 80 |
return SDG_NAMES
|
| 81 |
|
| 82 |
+
# Load Country Data
|
| 83 |
+
@st.cache_data
|
| 84 |
+
def getCountry():
|
| 85 |
+
# Read in countries from codelist
|
| 86 |
+
country_df = pd.read_csv('src/codelists/country_codes_ISO3166-1alpha-2.csv')
|
| 87 |
+
COUNTRY_CODES = country_df['Alpha-2 code'].tolist()
|
| 88 |
+
COUNTRY_NAMES = country_df['Country'].tolist()
|
| 89 |
+
|
| 90 |
+
COUNTRY_OPTION_LIST = [f"{COUNTRY_NAMES[i]} ({COUNTRY_CODES[i][-3:-1].upper()})"for i in range(len(COUNTRY_NAMES))]
|
| 91 |
+
|
| 92 |
+
return COUNTRY_OPTION_LIST
|
| 93 |
+
|
| 94 |
# Load Sentence Transformer Model
|
| 95 |
@st.cache_resource
|
| 96 |
def load_model():
|
|
|
|
| 122 |
CRS5_MERGED = getCRS5()
|
| 123 |
SDG_NAMES = getSDG()
|
| 124 |
|
| 125 |
+
COUNTRY_OPTION_LIST = getCountry()
|
| 126 |
+
|
| 127 |
model = load_model()
|
| 128 |
sentences, embeddings, faiss_index = load_embeddings_and_index()
|
| 129 |
|
|
|
|
| 167 |
|
| 168 |
|
| 169 |
with col2:
|
| 170 |
+
# COUNTRY SELECTION
|
| 171 |
+
|
| 172 |
+
|
| 173 |
+
country_option = st.multiselect(
|
| 174 |
+
'Country / Countries',
|
| 175 |
+
COUNTRY_OPTION_LIST,
|
| 176 |
+
placeholder="Select"
|
| 177 |
+
)
|
| 178 |
+
|
| 179 |
+
# ORGA SELECTION
|
| 180 |
+
orga_abbreviation = projects_df["orga_abbreviation"].unique()
|
| 181 |
+
orga_full_names = projects_df["orga_full_name"].unique()
|
| 182 |
+
orga_list = [f"{orga_full_names[i]} ({orga_abbreviation[i].upper()})"for i in range(len(orga_abbreviation))]
|
| 183 |
+
|
| 184 |
+
orga_option = st.multiselect(
|
| 185 |
+
'Development Bank / Organization',
|
| 186 |
+
orga_list,
|
| 187 |
+
placeholder="Select"
|
| 188 |
+
)
|
| 189 |
|
| 190 |
|
| 191 |
# CRS CODE LIST
|
|
|
|
| 198 |
else:
|
| 199 |
sdg_str = ""
|
| 200 |
|
| 201 |
+
# COUNTRY CODES LIST
|
| 202 |
+
country_code_list = [option[-3:-1] for option in country_option]
|
| 203 |
+
|
| 204 |
+
# ORGANIZATION CODES LIST
|
| 205 |
+
orga_code_list = [option.split("(")[1][:-1].lower() for option in orga_option]
|
| 206 |
+
|
| 207 |
# FILTER DF WITH SELECTED FILTER OPTIONS
|
| 208 |
+
filtered_df = filter_projects(projects_df, crs3_list, crs5_list, sdg_str, country_code_list, orga_code_list)
|
| 209 |
|
| 210 |
# FIND MATCHES
|
| 211 |
p1_df, p2_df = calc_matches(filtered_df, projects_df, sim_matrix)
|