-
Notifications
You must be signed in to change notification settings - Fork 3
/
vectors_demo.py
159 lines (125 loc) · 6.15 KB
/
vectors_demo.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
import math
from arango_queries import get_witnesses_for_hotspot
from pyArango.connection import Connection, Database
import os
from dotenv import load_dotenv
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
HOTSPOT_ADDRESS = "11RaVmAadLFRstHS5sZZ9Yr52nw3hTPoNaLwERx5Mjfrxa631Wh"
# HOTSPOT_ADDRESS = "11toz227sLe3xoA7vWgGt5ZnMQBDzuMdodYkcWpMFt9ERP5VfQq"
# HOTSPOT_ADDRESS = "112sLpwVU7Kx4pxBDFEAQPpatmELE8UmvXP25muCZQmykv4j2s6r"
# HOTSPOT_ADDRESS = "11RA5Q53xN2RpALQ5aRhi1ztxt1YoZG4CyUXL6t7mwgkHZDj9Av"
N = 3 # num nearest hotspots in vector search
load_dotenv()
try:
c = Connection(
arangoURL=os.getenv('ARANGO_URL'),
username=os.getenv('ARANGO_USERNAME'),
password=os.getenv('ARANGO_PASSWORD')
)
except ConnectionError:
raise Exception('Unable to connect to the ArangoDB instance. Please check that it is running and that you have supplied the correct URL/credentials in the .env file.')
db: Database = c['helium-graphs']
witnesses = get_witnesses_for_hotspot(db, HOTSPOT_ADDRESS)
witnesses = [w for w in witnesses if w["distance_m"] < 50000]
witnesses_df = pd.DataFrame(witnesses)
beacon_coords = np.array([w["coords_beacon"] for w in witnesses])
def calc_distance(pt1, pt2):
return np.sqrt((pt1[0] - pt2[0])**2 + (pt1[1] - pt2[1])**2)
def calc_theta(pt1, pt2):
return math.atan2(pt1[1] - pt2[1], pt1[0] - pt2[0])
def get_distance_matrices(witnesses):
distance_matrix = np.zeros((len(witnesses), len(witnesses)))
theta_matrix = np.zeros_like(distance_matrix)
for ii in range(len(witnesses)):
for jj in range(len(witnesses)):
if ii == jj:
distance_matrix[ii,jj] = np.nan
theta_matrix[ii,jj] = np.nan
continue
distance_matrix[ii,jj] = calc_distance(witnesses[ii]["coords_beacon"], witnesses[jj]["coords_beacon"])
theta_matrix[ii,jj] = calc_theta(witnesses[ii]["coords_beacon"], witnesses[jj]["coords_beacon"])
return distance_matrix, theta_matrix
distance_matrix, theta_matrix = get_distance_matrices(witnesses)
# fig, ax = plt.subplots()
net_dx, net_dy = np.zeros((len(witnesses),)), np.zeros((len(witnesses),))
# ax.scatter(beacon_coords[:,0], beacon_coords[:,1], alpha=0.3)
for i in range(len(witnesses)):
# ax.annotate(str(i), (beacon_coords[i,0], beacon_coords[i,1]))
# find nearest N vectors
nearest_idx = np.argpartition(distance_matrix[i,:], N)[:N]
theta = theta_matrix[i,nearest_idx]
r = [(witnesses[idx]["rssi"] - witnesses[i]["rssi"]) / (witnesses[idx]["rssi"] - witnesses[i]["distance_m"]) for idx in nearest_idx]
dx = [r[j] * math.cos(theta[j]) for j in range(len(theta))]
dy = [r[j] * math.sin(theta[j]) for j in range(len(theta))]
# for j in range(len(theta)):
# ax.arrow(beacon_coords[i,0], beacon_coords[i,1], dx[j], dy[j],
# color="gray", head_width=0.01)
net_dx[i], net_dy[i] = np.sum(dx), np.sum(dy)
net_theta = math.atan2(net_dy[i], net_dx[i])
# ax.arrow(beacon_coords[i,0], beacon_coords[i,1], net_dx[i], net_dy[i],
# color="orange", head_width=0.005)
# plt.show()
from scipy.interpolate import griddata
from haversine import haversine
x_steps = int(np.abs(max(beacon_coords[:,0]) - min(beacon_coords[:,0])) * 1000)
y_steps = int(np.abs(max(beacon_coords[:,1]) - min(beacon_coords[:,1])) * 1000)
grid_x, grid_y = np.meshgrid(np.linspace(min(beacon_coords[:,0]), max(beacon_coords[:,0]), x_steps),
np.linspace(min(beacon_coords[:,1]), max(beacon_coords[:,1]), y_steps))
grid_z = griddata(beacon_coords, [w["rssi"] for w in witnesses], (grid_x, grid_y), method="linear")
plt.title("RSSI")
plt.xlabel("Longitude")
plt.ylabel("Latitude")
plt.imshow(grid_z, origin="lower")
plt.colorbar()
plt.show()
grid_dx = griddata(beacon_coords, net_dx, (grid_x, grid_y), method="linear")
plt.imshow(grid_dx, origin="lower")
plt.title("dRSSI / dx")
plt.colorbar()
plt.show()
grid_dy = griddata(beacon_coords, net_dy, (grid_x, grid_y), method="linear")
plt.imshow(grid_dy, origin="lower")
plt.title("dRSSI / dy")
plt.colorbar()
plt.show()
H,W = grid_z.shape
grid_div = np.zeros_like(grid_z)
for i in range(1, H - 1):
for j in range(1, W - 1):
grid_div[i,j] = grid_dx[i,j+1] - grid_dx[i,j-1] + grid_dy[i+1,j] - grid_dy[i-1,j]
plt.imshow(grid_div, origin="lower")
plt.title("divergence")
plt.colorbar()
plt.show()
plt.pcolor(grid_x, grid_y, grid_div)
plt.xlabel("Longitude")
plt.ylabel("Latitude")
plt.show()
coords_idx_min_div = np.nanargmin(grid_div)
coords_idx_max_rssi = np.nanargmax(grid_z)
predicted_coords_div = (grid_y.flatten()[coords_idx_min_div], grid_x.flatten()[coords_idx_min_div])
predicted_coords_rssi = (grid_y.flatten()[coords_idx_max_rssi], grid_x.flatten()[coords_idx_max_rssi])
asserted_coords = (witnesses[0]['coords_witness'][1], witnesses[0]['coords_witness'][0])
print(f"Predicted coords (divergence): {predicted_coords_div}")
print(f"Distance from assert (divergence): {np.round(haversine(predicted_coords_div, asserted_coords), 3)} km")
print(f"Distance from assert (max rssi): {np.round(haversine(predicted_coords_rssi, asserted_coords), 3)} km")
fig, ax = plt.subplots()
ax.streamplot(grid_x, grid_y, grid_dx, grid_dy, color=np.sqrt(grid_dx**2+grid_dy**2), cmap="autumn")
# ax.quiver(grid_x, grid_y, grid_dx, grid_dy)
ax.scatter(asserted_coords[1], asserted_coords[0], marker="+", color="k", s=100)
ax.set_aspect("equal")
plt.show()
x_steps = int(np.abs(max(beacon_coords[:,0]) - min(beacon_coords[:,0])) * 50)
y_steps = int(np.abs(max(beacon_coords[:,1]) - min(beacon_coords[:,1])) * 50)
grid_x, grid_y = np.meshgrid(np.linspace(min(beacon_coords[:,0]), max(beacon_coords[:,0]), x_steps),
np.linspace(min(beacon_coords[:,1]), max(beacon_coords[:,1]), y_steps))
grid_dx = griddata(beacon_coords, net_dx, (grid_x, grid_y), method="linear")
grid_dy = griddata(beacon_coords, net_dy, (grid_x, grid_y), method="linear")
fig, ax = plt.subplots()
# ax.streamplot(grid_x, grid_y, grid_dx, grid_dy, color=np.sqrt(grid_dx**2+grid_dy**2), cmap="autumn")
ax.quiver(grid_x, grid_y, -grid_dx, -grid_dy)
ax.scatter(asserted_coords[1], asserted_coords[0], marker="+", color="k", s=100)
ax.set_aspect("equal")
plt.show()