Introduction to location intelligence In this research and study, we will analyze the food prices in Senegal for different cereals categories.
Introduction to location intelligence¶
In this research and study, we will analyze the food prices in Senegal for different cerales categories.
In [1]:
import pandas as pd
import numpy as np
import geopandas
import geopandas as gpd
from shapely.geometry import Point
import matplotlib
import matplotlib.pyplot as plt
import folium
import os
import seaborn as sns
In [2]:
shp_path = "D:/Research/PROJECT/pyexeriences/Geospatial experiences/Book Geographic data with python/data/food_price_regions_senegal/food_price_region_senegal.gpkg"
In [3]:
data = geopandas.read_file(shp_path)
d:\programm files\python 3 8 6\lib\site-packages\geopandas\geodataframe.py:422: RuntimeWarning: Sequential read of iterator was interrupted. Resetting iterator. This can negatively impact the performance. for feature in features_lst:
In [4]:
data.head()
Out[4]:
| field_1 | adm1id | address | city | country | category | catid | cmid | cmname | cmnameshor | ... | umid | unit | x | year | result_num | osm_id | display_na | type | latlong | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 2636 | Gueule Tapee | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 5 | KG | 1358208000 | 2013 | 0 | 560764203 | Gueule TapΓ©e, Gueule TapΓ©e-Fass-Colobane, Daka... | place | suburb | POINT (-17.45728 14.68287) |
| 1 | 1 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 5 | KG | 1360886400 | 2013 | 0 | 2447109686 | SGBS, Avenue Cheikh Ahmadou Bamba MbackΓ©, Derk... | amenity | bank | POINT (-17.44808 14.72122) |
| 2 | 1 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 5 | KG | 1360886400 | 2013 | 1 | 2447109687 | BOA, Avenue du PrΓ©sident Habib Bourguiba, CitΓ©... | amenity | bank | POINT (-17.44838 14.72078) |
| 3 | 2 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 5 | KG | 1363305600 | 2013 | 0 | 2447109686 | SGBS, Avenue Cheikh Ahmadou Bamba MbackΓ©, Derk... | amenity | bank | POINT (-17.44808 14.72122) |
| 4 | 2 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 5 | KG | 1363305600 | 2013 | 1 | 2447109687 | BOA, Avenue du PrΓ©sident Habib Bourguiba, CitΓ©... | amenity | bank | POINT (-17.44838 14.72078) |
5 rows × 30 columns
In [5]:
# Handling missing value
data.isna().sum()
Out[5]:
field_1 0 adm1id 0 address 0 city 0 country 0 category 0 catid 0 cmid 0 cmname 0 cmnameshor 0 currency 0 date 0 default 658 interpolat 0 label 0 mktid 0 price 0 ptid 0 scaling 0 sn 0 umid 0 unit 0 x 0 year 0 result_num 0 osm_id 0 display_na 0 type 0 latlong 0 geometry 0 dtype: int64
In [6]:
# Getting latitude and longitude of every city
data['latitude'] = data.geometry.y
data['longitude'] = data.geometry.x
In [7]:
data.head()
Out[7]:
| field_1 | adm1id | address | city | country | category | catid | cmid | cmname | cmnameshor | ... | x | year | result_num | osm_id | display_na | type | latlong | geometry | latitude | longitude | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 0 | 2636 | Gueule Tapee | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 1358208000 | 2013 | 0 | 560764203 | Gueule TapΓ©e, Gueule TapΓ©e-Fass-Colobane, Daka... | place | suburb | POINT (-17.45728 14.68287) | 14.682867 | -17.457283 |
| 1 | 1 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 1360886400 | 2013 | 0 | 2447109686 | SGBS, Avenue Cheikh Ahmadou Bamba MbackΓ©, Derk... | amenity | bank | POINT (-17.44808 14.72122) | 14.721220 | -17.448079 |
| 2 | 1 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 1360886400 | 2013 | 1 | 2447109687 | BOA, Avenue du PrΓ©sident Habib Bourguiba, CitΓ©... | amenity | bank | POINT (-17.44838 14.72078) | 14.720784 | -17.448379 |
| 3 | 2 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 1363305600 | 2013 | 0 | 2447109686 | SGBS, Avenue Cheikh Ahmadou Bamba MbackΓ©, Derk... | amenity | bank | POINT (-17.44808 14.72122) | 14.721220 | -17.448079 |
| 4 | 2 | 2636 | Castors | Dakar | Senegal | cereals and tubers | 1 | 158 | Maize (imported) - Retail | Maize imp. | ... | 1363305600 | 2013 | 1 | 2447109687 | BOA, Avenue du PrΓ©sident Habib Bourguiba, CitΓ©... | amenity | bank | POINT (-17.44838 14.72078) | 14.720784 | -17.448379 |
5 rows × 32 columns
In [8]:
data.columns
Out[8]:
Index(['field_1', 'adm1id', 'address', 'city', 'country', 'category', 'catid',
'cmid', 'cmname', 'cmnameshor', 'currency', 'date', 'default',
'interpolat', 'label', 'mktid', 'price', 'ptid', 'scaling', 'sn',
'umid', 'unit', 'x', 'year', 'result_num', 'osm_id', 'display_na',
'type', 'latlong', 'geometry', 'latitude', 'longitude'],
dtype='object')
We will select the most important columns within our data
In [9]:
dataset = data[['date', 'year', 'cmname', 'cmnameshor', 'category', 'price', 'address', 'city', 'country', 'longitude', 'latitude', 'geometry']]
In [10]:
dataset.head()
Out[10]:
| date | year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 1/15/2013 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) |
| 1 | 2/15/2013 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2 | 2/15/2013 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
| 3 | 3/15/2013 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 4 | 3/15/2013 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
In [11]:
dataset.columns
Out[11]:
Index(['date', 'year', 'cmname', 'cmnameshor', 'category', 'price', 'address',
'city', 'country', 'longitude', 'latitude', 'geometry'],
dtype='object')
In [12]:
dataset.shape
Out[12]:
(658, 12)
In [13]:
dataset.dtypes
Out[13]:
date object year object cmname object cmnameshor object category object price object address object city object country object longitude float64 latitude float64 geometry geometry dtype: object
In [14]:
# Change price type to float
dataset.price = dataset['price'].astype(np.float)
dataset.price.dtype
d:\programm files\python 3 8 6\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy super(GeoDataFrame, self).__setitem__(key, value)
Out[14]:
dtype('float64')
In [15]:
dataset.isna().sum() # Checking the missing value in all columns
Out[15]:
date 0 year 0 cmname 0 cmnameshor 0 category 0 price 0 address 0 city 0 country 0 longitude 0 latitude 0 geometry 0 dtype: int64
In [16]:
# convert the 'date' column to datetime
dataset['date'] = pd.to_datetime(dataset.date)
In [17]:
dataset.head()
Out[17]:
| date | year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | 2013-01-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) |
| 1 | 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2 | 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
| 3 | 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 4 | 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
In [18]:
# Total years
dataset.year.unique()
Out[18]:
array(['2013', '2014', '2015', '2016', '2017', '2018'], dtype=object)
The dataset analyze the total of 6 years from 2013 until to 2018 food price in different locations in Senegal.
In [19]:
(dataset['date'].max() - dataset['date'].min())
Out[19]:
Timedelta('2007 days 00:00:00')
The dataset analyze 2007 days almost 6 years mines holidays and weekends
In [20]:
2007/365 # A year is 365, our result divided with the number of year of the dataset
Out[20]:
5.498630136986302
The General Variation of Food Price¶
In [21]:
# To check the variation of the food price over time, let's make the date column as index
dataset.index = dataset['date']
In [22]:
dataset.head()
Out[22]:
| date | year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| date | ||||||||||||
| 2013-01-15 | 2013-01-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) |
| 2013-02-15 | 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2013-02-15 | 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
| 2013-03-15 | 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2013-03-15 | 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
In [23]:
dataset.columns
Out[23]:
Index(['date', 'year', 'cmname', 'cmnameshor', 'category', 'price', 'address',
'city', 'country', 'longitude', 'latitude', 'geometry'],
dtype='object')
In [24]:
# let's drop the date column
dataset.drop('date', axis=1, inplace=True)
d:\programm files\python 3 8 6\lib\site-packages\pandas\core\frame.py:4167: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy return super().drop(
In [25]:
dataset.head()
Out[25]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| date | |||||||||||
| 2013-01-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) |
In [26]:
dataset.tail()
Out[26]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| date | |||||||||||
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.024736 | 14.282493 | POINT (-15.02474 14.28249) |
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.548433 | 14.101970 | POINT (-15.54843 14.10197) |
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.547690 | 14.106511 | POINT (-15.54769 14.10651) |
| 2018-07-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 294.0 | Kaolack | Kaolack | Senegal | -15.788883 | 13.963176 | POINT (-15.78888 13.96318) |
| 2018-07-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 294.0 | Kaolack | Kaolack | Senegal | -16.076391 | 14.138815 | POINT (-16.07639 14.13881) |
It seems the data has been collected from day 15 on every month
In [27]:
# Check the unique price
plt.plot(dataset.price.unique())
Out[27]:
[<matplotlib.lines.Line2D at 0x1ca26a90>]
The above result is only useful when checking the highest and lowest price. Let's try to make the output more meaningful by sorting the values
In [28]:
unique_price = dataset.price.unique()
print(sorted(unique_price))
[175.0, 183.0, 187.0, 188.0, 190.0, 193.0, 200.0, 203.0, 205.0, 206.25, 206.6667, 207.0, 208.0, 210.0, 211.0, 211.25, 212.0, 212.5, 213.0, 213.3333, 213.75, 215.0, 217.0, 217.5, 218.0, 218.3333, 219.0, 220.0, 221.25, 221.6667, 222.5, 223.0, 223.3333, 225.0, 227.0, 227.5, 228.3333, 230.0, 231.0, 231.25, 231.6667, 233.3333, 235.0, 237.5, 240.0, 242.5, 245.0, 247.0, 248.4375, 250.0, 260.0, 268.0, 270.0, 271.6667, 272.5, 273.0, 273.3333, 273.75, 275.0, 280.0, 283.0, 285.0, 288.0, 290.0, 292.0, 294.0, 300.0, 450.0]
In [29]:
# It's also possible with dataframe
print(sorted(dataset['price'].unique()))
[175.0, 183.0, 187.0, 188.0, 190.0, 193.0, 200.0, 203.0, 205.0, 206.25, 206.6667, 207.0, 208.0, 210.0, 211.0, 211.25, 212.0, 212.5, 213.0, 213.3333, 213.75, 215.0, 217.0, 217.5, 218.0, 218.3333, 219.0, 220.0, 221.25, 221.6667, 222.5, 223.0, 223.3333, 225.0, 227.0, 227.5, 228.3333, 230.0, 231.0, 231.25, 231.6667, 233.3333, 235.0, 237.5, 240.0, 242.5, 245.0, 247.0, 248.4375, 250.0, 260.0, 268.0, 270.0, 271.6667, 272.5, 273.0, 273.3333, 273.75, 275.0, 280.0, 283.0, 285.0, 288.0, 290.0, 292.0, 294.0, 300.0, 450.0]
The highest price = 450 fcfa (The currancy is Franc cfa, West African Monetary CEDEAO except some countries, but Senegal is using this currancy) and the lowest price is 175
In [30]:
dataset[dataset.price == 450]
Out[30]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| date | |||||||||||
| 2014-06-15 | 2014 | Rice (ordinary, first quality) - Retail | Rice ordinary, 1st q. | cereals and tubers | 450.0 | Kedougou | Kedougou | Senegal | -12.185538 | 12.55709 | POINT (-12.18554 12.55709) |
In [31]:
dataset[dataset.price == 175]
Out[31]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | |
|---|---|---|---|---|---|---|---|---|---|---|---|
| date | |||||||||||
| 2015-12-15 | 2015 | Maize (local) - Retail | Maize loc. | cereals and tubers | 175.0 | Tambacounda | Tambacounda | Senegal | -13.089176 | 13.830153 | POINT (-13.08918 13.83015) |
| 2015-12-15 | 2015 | Maize (local) - Retail | Maize loc. | cereals and tubers | 175.0 | Tambacounda | Tambacounda | Senegal | -13.668290 | 13.769258 | POINT (-13.66829 13.76926) |
| 2015-12-15 | 2015 | Maize (local) - Retail | Maize loc. | cereals and tubers | 175.0 | Tambacounda | Tambacounda | Senegal | -13.672322 | 13.774000 | POINT (-13.67232 13.77400) |
| 2016-01-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Kolda | Kolda | Senegal | -14.939658 | 12.893608 | POINT (-14.93966 12.89361) |
| 2016-01-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Kolda | Kolda | Senegal | -14.142228 | 12.911966 | POINT (-14.14223 12.91197) |
| 2016-03-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Sedhiou | Sedhiou | Senegal | -15.557338 | 12.709905 | POINT (-15.55734 12.70990) |
| 2016-03-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Sedhiou | Sedhiou | Senegal | -15.551334 | 12.713448 | POINT (-15.55133 12.71345) |
| 2016-04-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Sedhiou | Sedhiou | Senegal | -15.557338 | 12.709905 | POINT (-15.55734 12.70990) |
| 2016-04-15 | 2016 | Millet - Retail | Millet | cereals and tubers | 175.0 | Sedhiou | Sedhiou | Senegal | -15.551334 | 12.713448 | POINT (-15.55133 12.71345) |
We can observe the higest price is in Rice in Kedougou which different from the price of other cereales categories, we just wanted to know the locations here
In [32]:
dataset['percentile'] = dataset.groupby(dataset.index)['price'].rank(pct=True)
d:\programm files\python 3 8 6\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy super(GeoDataFrame, self).__setitem__(key, value)
In [33]:
dataset.head()
Out[33]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | percentile | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| date | ||||||||||||
| 2013-01-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) | 0.833333 |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) | 0.700000 |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) | 0.700000 |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) | 0.687500 |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) | 0.687500 |
In [34]:
dataset.percentile.plot(figsize=(12, 10))
Out[34]:
<AxesSubplot:xlabel='date'>
In [35]:
dataset.percentile.hist()
Out[35]:
<AxesSubplot:>
In [36]:
y_percentile = dataset.groupby(dataset.index.year)['price'].rank(pct=True)
In [37]:
high_percentile = y_percentile.sort_values(ascending=False)
print('The high percentile over years: ', high_percentile[:10]);
print('The low percentile over years: ', high_percentile[-10:])
The high percentile over years: date 2014-06-15 1.000000 2017-11-15 1.000000 2018-06-15 0.992063 2018-07-15 0.992063 2013-02-15 0.987288 2013-04-15 0.987288 2013-03-15 0.987288 2013-03-15 0.987288 2017-07-15 0.982609 2017-07-15 0.982609 Name: price, dtype: float64 The low percentile over years: date 2016-01-15 0.029167 2014-01-15 0.026316 2014-02-15 0.026316 2014-04-15 0.026316 2014-05-15 0.026316 2014-06-15 0.026316 2014-07-15 0.026316 2015-12-15 0.018349 2015-12-15 0.018349 2015-12-15 0.018349 Name: price, dtype: float64
Variation of Price over month and year¶
In [38]:
dataset['month_variation'] = dataset.price.diff()
dataset['year_variation'] = dataset.price.diff(12)
d:\programm files\python 3 8 6\lib\site-packages\geopandas\geodataframe.py:853: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame. Try using .loc[row_indexer,col_indexer] = value instead See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy super(GeoDataFrame, self).__setitem__(key, value)
In [39]:
dataset
Out[39]:
| year | cmname | cmnameshor | category | price | address | city | country | longitude | latitude | geometry | percentile | month_variation | year_variation | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| date | ||||||||||||||
| 2013-01-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Gueule Tapee | Dakar | Senegal | -17.457283 | 14.682867 | POINT (-17.45728 14.68287) | 0.833333 | NaN | NaN |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) | 0.700000 | 0.0 | NaN |
| 2013-02-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) | 0.700000 | 0.0 | NaN |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448079 | 14.721220 | POINT (-17.44808 14.72122) | 0.687500 | 0.0 | NaN |
| 2013-03-15 | 2013 | Maize (imported) - Retail | Maize imp. | cereals and tubers | 250.0 | Castors | Dakar | Senegal | -17.448379 | 14.720784 | POINT (-17.44838 14.72078) | 0.687500 | 0.0 | NaN |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.024736 | 14.282493 | POINT (-15.02474 14.28249) | 0.555556 | 0.0 | -35.0 |
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.548433 | 14.101970 | POINT (-15.54843 14.10197) | 0.555556 | 0.0 | 0.0 |
| 2018-06-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 250.0 | Kaffrine | Kaffrine | Senegal | -15.547690 | 14.106511 | POINT (-15.54769 14.10651) | 0.555556 | 0.0 | 0.0 |
| 2018-07-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 294.0 | Kaolack | Kaolack | Senegal | -15.788883 | 13.963176 | POINT (-15.78888 13.96318) | 0.700000 | 44.0 | 44.0 |
| 2018-07-15 | 2018 | Sorghum - Retail | Sorghum | cereals and tubers | 294.0 | Kaolack | Kaolack | Senegal | -16.076391 | 14.138815 | POINT (-16.07639 14.13881) | 0.700000 | 0.0 | 44.0 |
658 rows × 14 columns
In [40]:
dataset.month_variation.unique()
Out[40]:
array([ nan, 0. , -7.5 , 7.5 , -20. , -10. ,
13.3333, -8.3333, 5.8333, 1.6667, -15. , 21.6667,
1.6666, -2.0833, -11.25 , 20. , 7. , -2. ,
-23. , 13. , -5. , 5. , -12. , 12. ,
-7. , -18. , -1.6667, 4.1667, -17.5 , 10. ,
-25. , -35. , 15. , 3. , -6. , 13.75 ,
-2.75 , 14. , 6. , -28. , 16. , -19. ,
17. , 8. , 22. , -27. , -13. , 12.5 ,
-2.5 , 3.3333, -3.3333, -16.25 , 0.4167, -6.6667,
6.25 , 3.75 , 2. , -17. , -8. , 25. ,
18. , 23. , -44. , 100. , -1.25 , -3.75 ,
0.8333, 2.5 , -1.5625, 1.5625, 175. , -180. ,
-47.5 , -13.75 , 12.0833, 11.6667, 6.6667, -13.3333,
-9.1667, 33. , 44. ])
In [41]:
dataset.year_variation.unique()
Out[41]:
array([ nan, -20. , -30. , -16.6667, -25. , -19.1667,
-17.5 , -5. , 5. , -23.3333, 6.6667, 14.1667,
10. , 7.9167, -10. , -11.6667, -13.3333, -11.25 ,
0. , 7. , 27. , 25. , 3. , -7. ,
-12. , 2. , 12. , 17. , 3.3333, 7.5 ,
-3. , -13. , -27.5 , -15. , 20. , -35. ,
-17. , -23. , 23.75 , 21. , 32. , 38. ,
44. , 19. , -13.75 , -11. , -31. , 14. ,
6. , 1. , 47. , 13. , -47. , -27. ,
37.5 , 15. , -1.6667, -3.3333, 1.6667, 27.5 ,
12.5 , -3.75 , -18.75 , -18.3333, -45. , -37.5 ,
-31.25 , 5.75 , -1.25 , -6.6667, -37. , 50. ,
18. , 43. , 73. , -50. , -85. , -87. ,
-75. , 100. , 75. , 73.75 , 85.75 , 82. ,
70. , 71.6667, -28.3333, -2.5 , 8. , -2. ,
-40. , -1.5625, 1.5625, -8.4375, 30. , 175. ,
-47.5 , -70. , -220. , 17.5 , 2.5 , -12.5 ,
-26.25 , -8.75 , -21.6667, 13.75 , 38.75 , -1.6666,
8.3334, -5.8333, 28.3333, 65. , 63. , 58. ,
60. , -33. ])
In [42]:
# The variation over the year
var_over_year = dataset.groupby(dataset.index.year)['price'].diff()
In [43]:
var_over_year.plot()
Out[43]:
<AxesSubplot:xlabel='date'>
Between the year of 2014 and 2015 there is a big changes in price happening. But the question is: In which cerales categories and in which region, we can go over and try to reloct ate the reasons
In [44]:
sns.lineplot(data=dataset, x="year", y="price")
Out[44]:
<AxesSubplot:xlabel='year', ylabel='price'>
Let's find the most important cities for food business with seaborn
In [45]:
sns.lineplot(data=dataset, x="year", y="price", hue= dataset.city)
Out[45]:
<AxesSubplot:xlabel='year', ylabel='price'>
In [46]:
sns.lineplot(
data=dataset.query("city == 'Dakar'"),
x="year", y="cmnameshor", hue="price",
estimator=None, lw=1,
)
Out[46]:
<AxesSubplot:xlabel='year', ylabel='cmnameshor'>
In [47]:
sum_price = dataset.groupby('year')['cmname', 'price'].sum()
sum_price
<ipython-input-47-1768111d0136>:1: FutureWarning: Indexing with multiple keys (implicitly converted to a tuple of keys) will be deprecated, use a list instead.
sum_price = dataset.groupby('year')['cmname', 'price'].sum()
Out[47]:
| price | |
|---|---|
| year | |
| 2013 | 28717.5000 |
| 2014 | 31693.2291 |
| 2015 | 24940.8334 |
| 2016 | 27398.7500 |
| 2017 | 27891.0000 |
| 2018 | 15095.0000 |
In [48]:
plt.plot(sum_price)
Out[48]:
[<matplotlib.lines.Line2D at 0x1f01ac70>]
The price has been dropped drammatically from 2017 to 2018
Let's find the most wanted food in Senegal
In [49]:
price_crops = dataset.dissolve('cmnameshor')['price'].sort_values(ascending=False)
In [50]:
# The most wanted cereales in Senegal
price_crops
Out[50]:
cmnameshor Rice ordinary, 1st q. 450.0000 Rice imp. 300.0000 Rice loc. 275.0000 Rice ordinary, 2nd q. 270.0000 Maize imp. 250.0000 Millet 225.0000 Maize loc. 223.3333 Sorghum 222.5000 Name: price, dtype: float64
In [51]:
plt.hist(price_crops)
Out[51]:
(array([3., 1., 2., 1., 0., 0., 0., 0., 0., 1.]),
array([222.5 , 245.25, 268. , 290.75, 313.5 , 336.25, 359. , 381.75,
404.5 , 427.25, 450. ]),
<BarContainer object of 10 artists>)
Visualizing more on the price¶
In [52]:
dataset[['year','cmnameshor', 'price']].plot(figsize=(10,10))
Out[52]:
<AxesSubplot:xlabel='date'>
In [53]:
dataset['percentile'].hist(bins = 50, figsize = (10,5))
Out[53]:
<AxesSubplot:>
In [54]:
dataset.percentile.describe()
Out[54]:
count 658.000000 mean 0.550152 std 0.276764 min 0.066667 25% 0.301923 50% 0.538462 75% 0.791667 max 1.000000 Name: percentile, dtype: float64
We can observe from the figure above the concentration of main cities and local densities in the western side of the country along to the center-west
Commercial Zones:¶
The best location for food commercialization
In [55]:
# let’s find the most important city for food commercialization
fig, ax = plt.subplots(figsize=(10,10))
plot = sns.countplot(y = 'address',
color='grey',
order = dataset['address'].value_counts().index,
data=dataset, ax=ax)
plot.figure.savefig('food_price_senegal.png')
plot
plt.tight_layout()
plt.show()
In [56]:
# High to low demand on cities
dataset.city.value_counts()
Out[56]:
Dakar 125 Diourbel 93 Thies 93 Kolda 82 Kaolack 56 Sedhiou 50 Kaffrine 43 Tambacounda 41 Kedougou 31 Louga 19 Fatick 18 Saint louis 5 Matam 2 Name: city, dtype: int64
In [57]:
dataset.city.value_counts().plot()
Out[57]:
<AxesSubplot:>
In [58]:
# Mapping price over cities
dataset.plot(column = 'city', scheme='quantiles', cmap='Pastel2', figsize=(10,10))
Out[58]:
<AxesSubplot:>
Let's map the locations of our data into Senegal map
In [59]:
world = geopandas.read_file(geopandas.datasets.get_path('naturalearth_lowres'))
world.head()
Out[59]:
| pop_est | continent | name | iso_a3 | gdp_md_est | geometry | |
|---|---|---|---|---|---|---|
| 0 | 920938 | Oceania | Fiji | FJI | 8374.0 | MULTIPOLYGON (((180.00000 -16.06713, 180.00000... |
| 1 | 53950935 | Africa | Tanzania | TZA | 150600.0 | POLYGON ((33.90371 -0.95000, 34.07262 -1.05982... |
| 2 | 603253 | Africa | W. Sahara | ESH | 906.5 | POLYGON ((-8.66559 27.65643, -8.66512 27.58948... |
| 3 | 35623680 | North America | Canada | CAN | 1674000.0 | MULTIPOLYGON (((-122.84000 49.00000, -122.9742... |
| 4 | 326625791 | North America | United States of America | USA | 18560000.0 | MULTIPOLYGON (((-122.84000 49.00000, -120.0000... |
In [60]:
senegal = world[world.name == 'Senegal']
In [61]:
fig, ax = plt.subplots(figsize=(12,10))
senegal.plot(ax=ax, edgecolor = 'black', color='white', figsize=(12, 10))
dataset.plot(markersize=20, ax=ax);
plt.savefig('food_price_map_senegal.png', bbox_inches='tight')
In [62]:
import contextily as ctx
print(ctx.providers.keys())
dict_keys(['OpenStreetMap', 'OpenSeaMap', 'OpenPtMap', 'OpenTopoMap', 'OpenRailwayMap', 'OpenFireMap', 'SafeCast', 'Thunderforest', 'OpenMapSurfer', 'Hydda', 'MapBox', 'Stamen', 'Esri', 'OpenWeatherMap', 'HERE', 'FreeMapSK', 'MtbMap', 'CartoDB', 'HikeBike', 'BasemapAT', 'nlmaps', 'NASAGIBS', 'NLS', 'JusticeMap', 'Wikimedia', 'GeoportailFrance', 'OneMapSG'])
In [63]:
print(ctx.providers.OpenStreetMap.keys())
dict_keys(['Mapnik', 'DE', 'CH', 'France', 'HOT', 'BZH'])
In [64]:
# add open street map basemap
ctx.add_basemap(ax, crs = dataset.crs, url = ctx.providers.OpenStreetMap.Mapnik)
<ipython-input-64-aa1f9c3384c2>:2: FutureWarning: The "url" option is deprecated. Please use the "source" argument instead. ctx.add_basemap(ax, crs = dataset.crs, url = ctx.providers.OpenStreetMap.Mapnik)
<Figure size 432x288 with 0 Axes>
In [65]:
fig, ax = plt.subplots(figsize=(14,12))
senegal.plot(ax=ax, color='None', edgecolor='black')
dataset.plot(ax=ax)
ax.set(title='food price in Senegal') # add a title to the map
ax.set_axis_off() # remove the axis ticks
plt.tight_layout() # adjust the padding between figure edges
plt.savefig('foodpricesenegal_conf_OSM_Map.png')# Save the map as an image
ctx.add_basemap(ax, crs=dataset.crs.to_string())
What we can understand from above map is the shape of Senegal from geopandas dataset in not well fitting with the map from contextily available map datasets
In [66]:
# The locations in Dakar, the main city and capital of Senegal
sen = dataset.query("city == 'Dakar'")
ax = sen.plot(facecolor="none",
edgecolor="red",
linewidth=2, figsize=(12, 10)
)
ctx.add_basemap(ax,
crs=sen.crs.to_string(),
source=ctx.providers.CartoDB.Voyager
)
In [67]:
ax = dataset.plot(facecolor="none",
edgecolor="red",
linewidth=2, figsize=(10, 10)
)
ctx.add_basemap(ax,
crs=sen.crs.to_string(),
source=ctx.providers.CartoDB.Voyager
)
Mapping with Folium Library
In [68]:
m = folium.Map (
location = [14.414808661893874, -14.506641213432777],
tiles='Stamen Toner',
zoom_start = 8
)
folium.GeoJson(data).add_to(m)
m
Out[68]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [69]:
fig, ax = plt.subplots(figsize=(12,10))
dataset.plot(markersize=5,ax=ax);
#ax.set_xticklabels([])
#ax.set_yticklabels([])
ax.set_title('Food price in different locations in Senegal')
plt.show()
In [70]:
data.cmname.unique() # unique value for food categories
Out[70]:
array(['Maize (imported) - Retail', 'Maize (local) - Retail',
'Millet - Retail', 'Rice (imported) - Retail',
'Rice (local) - Retail', 'Rice (ordinary, first quality) - Retail',
'Rice (ordinary, second quality) - Retail', 'Sorghum - Retail'],
dtype=object)
In [71]:
data.cmnameshor.unique()
Out[71]:
array(['Maize imp.', 'Maize loc.', 'Millet', 'Rice imp.', 'Rice loc.',
'Rice ordinary, 1st q.', 'Rice ordinary, 2nd q.', 'Sorghum'],
dtype=object)
In [72]:
local_rice = data[data['cmnameshor']== 'Rice loc.']
In [73]:
imp_rice = data[data['cmnameshor']== 'Rice imp.']
In [74]:
fig, ax = plt.subplots(figsize=(10,10))
local_rice.plot(ax=ax)
imp_rice.plot(ax=ax)
plt.show()
The imported rice is much more consomable in Senegal than the local cultivated rice.
In [75]:
folium.GeoJson(local_rice).add_to(m)
folium.GeoJson(imp_rice, popup='Imported rice').add_to(m)
m
Out[75]:
Make this Notebook Trusted to load map: File -> Trust Notebook
In [76]:
from folium.plugins import FastMarkerCluster
lons = data['longitude']
lats = data['latitude']
m1 = folium.Map(
location = [np.mean(data.latitude), np.mean(data.longitude)],
tiles= 'Stamen Toner',
zoom_start=6
)
FastMarkerCluster(data=list(zip(lats, lons))).add_to(m1)
folium.LayerControl().add_to(m1)
m1
Out[76]:
Make this Notebook Trusted to load map: File -> Trust Notebook
Comments
Post a Comment