-
Notifications
You must be signed in to change notification settings - Fork 19
/
TSPGraph.py
230 lines (184 loc) · 8.7 KB
/
TSPGraph.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
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
import numpy as np
import matplotlib.pyplot as plt
"""
Based on:
https://github.com/notadamking/Stock-Trading-Visualization
"""
class TSPGraph:
"""
TSP rendering for the TSP environment
"""
def __init__(self, window_size=10, time=0):
"""
Initialise TSP matplotlib Graphs
:param int window_size: window size of the distances
"""
self.window_size = window_size
# create a figure on screen and set the title
fig = plt.figure()
fig.suptitle('TSP')
# create top subplot for net worth axis
self.best_distance_ax = plt.subplot2grid((12, 1),
(0, 0),
rowspan=2,
colspan=1)
# create bottom subplot for TSP plots
self.tour_ax = plt.subplot2grid((12, 1),
(3, 0),
rowspan=4,
colspan=1)
# create 2nd bottom subplot for TSP plots
self.best_tour_ax = plt.subplot2grid((12, 1),
(8, 0),
rowspan=4,
colspan=1)
# add padding to make graph easier to view
plt.subplots_adjust(left=0.11,
bottom=0.24,
right=0.90,
top=0.90,
wspace=0.2,
hspace=1)
def _render_distances(self, step_range):
"""
Render distances
:param list step_range: Steps to consider from the episode
"""
# clear the frame rendered last step
self.best_distance_ax.clear()
self.best_distances = [x/10000 for x in self.best_distances]
self.current_distances = [x/10000 for x in self.current_distances]
# plot the distances
self.best_distance_ax.plot(step_range,
self.best_distances,
'-', label='Best Distance',
color = "darkred")
self.best_distance_ax.plot(step_range,
self.current_distances,
'-', label='Current Distance', color="black")
# show legend, which uses the labels defined above
self.best_distance_ax.legend()
legend = self.best_distance_ax.legend(loc=0, ncol=2, prop={'size': 8})
legend.get_frame().set_alpha(0.4)
# save the last seen distance and the last step
last_distance = self.current_distances[-1]
best_distance = self.best_distances[-1]
last_step = self.step_range[-1]
# annotate the current distance on the graph
self.best_distance_ax.annotate('{0:.2f}'.format(last_distance),
(last_step, last_distance),
xytext=(last_step, last_distance),
bbox=dict(boxstyle='round',
fc='w', ec='k', lw=1),
color="black",
fontsize=8)
# annotate the current distance on the graph
self.best_distance_ax.annotate('{0:.2f}'.format(best_distance),
(last_step, best_distance),
xytext=(last_step, best_distance),
bbox=dict(boxstyle='round',
fc='w', ec='k', lw=1),
color="darkred",
fontsize=8)
# Add space above "max distance"
self.best_distance_ax.set_ylim(0,
max(np.maximum(self.best_distances,
self.current_distances) * 2))
# plt.show(block=False)
# self.fig.canvas.draw()
def _render_tour(self, positions, axis):
"""
Render TSP solutions in 2D
:param np.array positions: Positions of (tour_len, 2) points
"""
# clear the frame rendered last step
# self.tour_ax.clear()
axis.clear()
# transform positions to np array
# solution is eq. to traversing the graph in order
solution = [x for x in range(positions.shape[0])]
# plot scatters - red: depot
# self.tour_ax.scatter(positions[:, 0], positions[:, 1])
# self.tour_ax.scatter(positions[0, 0], positions[0, 1], color='darkred')
axis.scatter(positions[:, 0], positions[:, 1])
axis.scatter(positions[0, 0], positions[0, 1], color='darkred')
#plot segments connecting the nodes and calculate tour distance
start_node = 0
distance = 0.
N = len(solution)
for i in range(N-1):
start_pos = positions[start_node]
next_node = solution[i + 1]
end_pos = positions[next_node]
# self.tour_ax.annotate("",
# xy=start_pos, xycoords='data',
# xytext=end_pos, textcoords='data',
# arrowprops=dict(arrowstyle="-",
# connectionstyle="arc3"))
axis.annotate("",
xy=start_pos, xycoords='data',
xytext=end_pos, textcoords='data',
arrowprops=dict(arrowstyle="-",
connectionstyle="arc3"))
distance += np.linalg.norm(end_pos - start_pos)
start_node = next_node
# self.tour_ax.annotate("",
# xy=positions[start_node], xycoords='data',
# xytext=positions[0], textcoords='data',
# arrowprops=dict(arrowstyle="-",
# connectionstyle="arc3"))
axis.annotate("",
xy=positions[start_node], xycoords='data',
xytext=positions[0], textcoords='data',
arrowprops=dict(arrowstyle="-",
connectionstyle="arc3"))
distance += np.linalg.norm(positions[start_node] - positions[0])
textstr = "N nodes: %d\nTotal length: %.2f" % (N, distance)
props = dict(boxstyle='round', facecolor='wheat', alpha=0.5)
# self.tour_ax.text(0.05, 0.95,
# textstr,
# transform=self.tour_ax.transAxes,
# fontsize="small",
# verticalalignment='top',
# bbox=props)
axis.text(0.05, 0.95,
textstr,
transform=axis.transAxes,
fontsize="small",
verticalalignment='top',
bbox=props)
def render(self,
current_step,
best_distances,
current_distances,
state,
best_state):
"""
Render TSP distances and solutions in 2D
:param int current_step: current episode step
:param list best_distances: best distances found in an episode
:param list current_distances: current distances in an episode
:param torch.tensor state: positions of (tour_len, 2) points
"""
self.best_distances = np.zeros(self.window_size)
self.current_distances = np.zeros(self.window_size)
window_start = max(current_step - self.window_size, 0)
self.step_range = range(window_start, current_step + 1)
s = slice(window_start, current_step + 1)
self.best_distances = best_distances[s]
self.current_distances = current_distances[s]
self._render_distances(self.step_range)
self._render_tour(state, self.tour_ax)
self._render_tour(best_state, self.best_tour_ax)
self.tour_ax.set_title('Current Tour', fontsize=8)
self.best_tour_ax.set_title('Best Tour', fontsize=8)
# hide duplicate labels
self.tour_ax.get_xaxis().set_visible(False)
plt.setp(self.best_distance_ax.get_xticklabels(), visible=False)
# self.fig.canvas.draw()
# Necessary to view frames before they are unrendered
# plt.show(block=False)
plt.pause(0.01)
# show the graph without blocking the rest of the program
def close(self):
plt.close()