Table of Contents

• 1  Build Matrix

Build Matrix¶

In [11]:

import geopandas as gpd
import pandas as pd
import numpy as np
import os
import numpy.linalg as la
from sklearn.preprocessing import normalize

import geopandas as gpd import pandas as pd import numpy as np import os import numpy.linalg as la from sklearn.preprocessing import normalize

In [12]:

def pageRank(linkMatrix, d) :
    n = linkMatrix.shape[0]
    M = d * linkMatrix + (1-d)/n * np.ones([n, n]) # np.ones() is the J matrix, with ones for each entry.
    r = 100 * np.ones(n) / n # Sets up this vector (6 entries of 1/6 × 100 each)
    lastR = r
    r = M @ r
    i = 0
    while la.norm(lastR - r) > 0.01 :
        lastR = r
        r = M @ r
        i += 1
    return r

def pageRank(linkMatrix, d) : n = linkMatrix.shape[0] M = d * linkMatrix + (1-d)/n * np.ones([n, n]) # np.ones() is the J matrix, with ones for each entry. r = 100 * np.ones(n) / n # Sets up this vector (6 entries of 1/6 × 100 each) lastR = r r = M @ r i = 0 while la.norm(lastR - r) > 0.01 : lastR = r r = M @ r i += 1 return r

In [3]:

grid_path = "../Data/Fishnet500/Net500m.shp"
gdf = gpd.read_file(grid_path)
gdf.head()

grid_path = "../Data/Fishnet500/Net500m.shp" gdf = gpd.read_file(grid_path) gdf.head()

Out[3]:

	Id	Shape_Leng	Shape_Area	geometry
0	0	0.02079	2.701401e-05	POLYGON ((-97.65444 30.50675, -97.65964 30.506...
1	0	0.00750	9.236718e-07	POLYGON ((-97.57298 30.34562, -97.57648 30.344...
2	0	0.02079	2.701401e-05	POLYGON ((-97.96109 30.34043, -97.96629 30.340...
3	0	0.02079	2.701400e-05	POLYGON ((-97.94030 30.37681, -97.94550 30.376...
4	0	0.02079	2.701401e-05	POLYGON ((-97.72201 30.41319, -97.72720 30.413...

In [4]:

end_df_lst = os.listdir('end_pt')
end_df_lst

end_df_lst = os.listdir('end_pt') end_df_lst

Out[4]:

['RideAustin_0_3.csv',
 'RideAustin_12_15.csv',
 'RideAustin_15_18.csv',
 'RideAustin_18_21.csv',
 'RideAustin_21_24.csv',
 'RideAustin_3_6.csv',
 'RideAustin_6_9.csv',
 'RideAustin_9_12.csv']

In [28]:

%%time

rows = gdf.shape[0]

output_as_point = True

pt_lst = [x.centroid for x in gdf.geometry]
gdf_new = gpd.GeoDataFrame(gdf, geometry = pt_lst)

for df in end_df_lst:
    matrix = np.zeros((rows,rows))
    start_df = pd.read_csv(os.path.join('start_pt',df))
    end_df = pd.read_csv(os.path.join('end_pt',df))
    tdf = start_df.merge(end_df, left_on = 'Pid',right_on = 'Pid')
    print(f'Start df : {start_df.shape[0]}, End df : {end_df.shape[0]}, Total : {tdf.shape[0]}')
    try:
        for row in tdf.iloc:
            matrix[row.Gid_y][row.Gid_x] +=1
        matrix_norm = normalize(matrix, axis=0, norm='l1')
        r = pageRank(matrix_norm,0.9)
        if output_as_point:
            gdf_new['Rank'] = r
            fn = os.path.join('../Data/Output',str(df).replace('.csv','_pt.shp'))
            gdf_new.to_file(fn)         
        else:    
            gdf['Rank'] = r
            fn = os.path.join('../Data/Output',str(df).replace('.csv','.shp'))
            gdf.to_file(fn)
        print(f'Complete {fn}')
    except Exception as e:
        print(e)

%%time rows = gdf.shape[0] output_as_point = True pt_lst = [x.centroid for x in gdf.geometry] gdf_new = gpd.GeoDataFrame(gdf, geometry = pt_lst) for df in end_df_lst: matrix = np.zeros((rows,rows)) start_df = pd.read_csv(os.path.join('start_pt',df)) end_df = pd.read_csv(os.path.join('end_pt',df)) tdf = start_df.merge(end_df, left_on = 'Pid',right_on = 'Pid') print(f'Start df : {start_df.shape[0]}, End df : {end_df.shape[0]}, Total : {tdf.shape[0]}') try: for row in tdf.iloc: matrix[row.Gid_y][row.Gid_x] +=1 matrix_norm = normalize(matrix, axis=0, norm='l1') r = pageRank(matrix_norm,0.9) if output_as_point: gdf_new['Rank'] = r fn = os.path.join('../Data/Output',str(df).replace('.csv','_pt.shp')) gdf_new.to_file(fn) else: gdf['Rank'] = r fn = os.path.join('../Data/Output',str(df).replace('.csv','.shp')) gdf.to_file(fn) print(f'Complete {fn}') except Exception as e: print(e)

Start df : 95296, End df : 95296, Total : 95296
Complete ../Data/Output\RideAustin_0_3_pt.shp
Start df : 25794, End df : 25794, Total : 25794
Complete ../Data/Output\RideAustin_12_15_pt.shp
Start df : 45121, End df : 35139, Total : 35138
only integers, slices (`:`), ellipsis (`...`), numpy.newaxis (`None`) and integer or boolean arrays are valid indices
Start df : 49425, End df : 49425, Total : 49425
Complete ../Data/Output\RideAustin_18_21_pt.shp
Start df : 60570, End df : 60570, Total : 60570
Complete ../Data/Output\RideAustin_21_24_pt.shp
Start df : 100151, End df : 100151, Total : 100151
Complete ../Data/Output\RideAustin_3_6_pt.shp
Start df : 71477, End df : 77979, Total : 71477
Complete ../Data/Output\RideAustin_6_9_pt.shp
Start df : 18060, End df : 18060, Total : 18060
Complete ../Data/Output\RideAustin_9_12_pt.shp
Wall time: 41 s

In [6]:

tdf.head()

tdf.head()

Out[6]:

	Pid	Gid_x	Gid_y
0	7941	5541	5541
1	7942	2238	4559
2	7943	1238	4947
3	7944	4350	2479
4	7945	1238	4940

In [13]:

matrix_norm = normalize(matrix, axis=0, norm='l1')
matrix_norm

matrix_norm = normalize(matrix, axis=0, norm='l1') matrix_norm

Out[13]:

array([[0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       ...,
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.],
       [0., 0., 0., ..., 0., 0., 0.]])

In [14]:

r = pageRank(matrix_norm,0.9)
r

r = pageRank(matrix_norm,0.9) r

Out[14]:

array([1.15521465e-05, 1.15521465e-05, 1.15521465e-05, ...,
       1.15521465e-05, 1.15521465e-05, 1.15521465e-05])

In [ ]:

gdf['Rank'] = r
gdf.head()

gdf['Rank'] = r gdf.head()

In [ ]:

gdf.to_file('../Data/Output/RideAustin_6_8.shp')

gdf.to_file('../Data/Output/RideAustin_6_8.shp')

In [ ]: