/
experiments.py
287 lines (258 loc) · 10.6 KB
/
experiments.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
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
from timeit import default_timer as timer
import itertools
import generatebids as gb
import productmix as pm
import os
import csv
import gc
gc.disable() # Makes tests more reliable
def generate_file(experiment_name, n, m, M, q, rep):
filename = ("experiments/{0}/experiment-{1}-{2}-{3}-{4}-{5}.json"
.format(experiment_name, n, m, M, q, rep))
print('Generating file {}'.format(filename))
alloc = gb.get_allocation_problem(n, m, M, q)
pm.save_to_json(alloc, filename)
def generate_experiment_files(experiment_name, n_list, m_list, M_list, q_list,
reps=10, number=1):
# Generate folder 'experiment_name' if it does not exist
directory = "experiments/{}".format(experiment_name)
if not os.path.exists(directory):
os.makedirs(directory)
# Generate all experimental allocation files
for n, m, M, q in itertools.product(n_list, m_list, M_list, q_list):
for r in reps:
generate_file(experiment_name, n, m, M, q, r)
def time_unit_min_up(filename, M, number=1):
total_time = 0
for _ in range(number):
alloc = pm.load_from_json(filename)
start = timer()
_, steps = pm.min_up(alloc, long_step_routine="", test=True)
end = timer()
total_time += end-start
avg_time = total_time / number
return avg_time, steps
def time_binary_min_up(filename, M, number=1):
total_time = 0
for _ in range(number):
alloc = pm.load_from_json(filename)
start = timer()
_, steps = pm.min_up(alloc, long_step_routine="binarysearch", test=True)
end = timer()
total_time += end-start
avg_time = total_time / number
return avg_time, steps
def time_demand_min_up(filename, M, number=1):
total_time = 0
for _ in range(number):
alloc = pm.load_from_json(filename)
start = timer()
_, steps = pm.min_up(alloc, long_step_routine="demandchange", test=True)
end = timer()
total_time += end-start
avg_time = total_time / number
return avg_time, steps
def time_alloc(filename, M, number=1):
total_time = 0
for _ in range(number):
alloc = pm.load_from_json(filename)
gb.set_market_clearing_prices(alloc, M)
start = timer()
_, proc1, proc2, proc3 = pm.allocate(alloc, test=True)
end = timer()
total_time += end-start
avg_time = total_time / number
return avg_time, proc1, proc2, proc3
def avg(outputs):
"""Takes as input a list of output tuples and returns the element-wise
average.
"""
return [sum(o)/len(o) for o in zip(*outputs)]
def run_test(experiment, time_routine):
all_outputs = []
experiment_name, n_list, m_list, M_list, q_list, reps, number = experiment
for n, m, M, q in itertools.product(n_list, m_list, M_list, q_list):
out_vals = []
for r in reps:
filename = ("experiments/{0}/experiment-{1}-{2}-{3}-{4}-{5}.json"
.format(experiment_name, n, m, M, q, r))
print("Running file {}".format(filename))
while True:
try:
out_vals.append(time_routine(filename, M, number=number))
break
except ValueError:
print("Regenerating file due to ValueError")
generate_file(experiment_name, n, m, M, q, r)
all_outputs.append(avg(out_vals)) # average the times for reps
return zip(*all_outputs)
def get_data(experiment_name, f, i, j):
filename = 'experiments/csv/{}-{}.csv'.format(experiment_name, f)
with open(filename) as csvfile:
reader = csv.reader(csvfile, delimiter=',')
x_coords, y_coords = [], []
for row in reader:
x_coords.append(row[i])
y_coords.append(row[j])
return x_coords, y_coords
def visualise(experiment_name, f, x_col, y_col):
import matplotlib.pyplot as plt
x_coords, y_coords = get_data(experiment_name, f, x_col, y_col)
plt.clf()
plt.plot(x_coords, y_coords)
plt.ylabel('seconds')
filename = 'experiments/figures/{}-{}.png'.format(experiment_name, experiment_type)
plt.savefig(filename)
def save_to_file(exp_name, exp_type, *columns):
filename = 'experiments/csv/{}-{}.csv'.format(exp_name, exp_type)
with open(filename, 'w') as csvfile:
writer = csv.writer(csvfile, delimiter=',')
writer.writerows(zip(*columns))
def test_binary_min_up(experiment):
return run_test(experiment, time_routine=time_binary_min_up)
def test_unit_min_up(experiment):
return run_test(experiment, time_routine=time_unit_min_up)
def test_demand_min_up(experiment):
return run_test(experiment, time_routine=time_demand_min_up)
def test_allocate(experiment):
return run_test(experiment, time_routine = time_alloc)
def test_allocate(experiment):
return run_test(experiment, time_routine = time_alloc)
def run_minup_tests(experiment, i):
"""Runs MinUp with unit steps, binary search and demand change techniques.
Saves csv file with entries:
unit_times, unit_steps,
binary_times, binary_steps,
demand_times, demand_steps
"""
print("Starting {}".format(experiment[0]))
# Run tests
unit_times, unit_steps = test_unit_min_up(experiment)
binary_times, binary_steps = test_binary_min_up(experiment)
demand_times, demand_steps = test_demand_min_up(experiment)
# Save to CSV file
save_to_file(experiment[0], 'minup', experiment[i],
unit_times, unit_steps,
binary_times, binary_steps,
demand_times, demand_steps)
def run_allocate_test(experiment, i):
allocate_times, proc1, proc2, proc3 = test_allocate(experiment)
# Save to CSV file
save_to_file(experiment[0], 'allocate', experiment[i],
allocate_times, proc1, proc2, proc3)
if __name__ == "__main__":
### EXPERIMENTS
# control variables
generate = True
allocate = True
plot = True
# variables
large_n = [10]
small_n = [2]
n_range = range(5, 51, 5)
small_n_range = range(2,15)
q_range = range(20, 501, 20)
large_M = [10000]
small_M = [100]
large_q = [100]
small_q = [50]
m = [5]
m_range = range(2,21)
reps = range(0,50)
number = 1
### DEFINE EXPERIMENTS
experiment8 = ('experiment8', small_n, m, small_M, q_range, reps, 10)
experiment9 = ('experiment9', large_n, m, small_M, q_range, reps, 10)
experiment10 = ('experiment10', n_range, m, small_M, small_q, reps, 2)
experiment11 = ('experiment11', n_range, m, small_M, large_q, reps, 2)
experiment12 = ('experiment12',small_n,m_range,small_M,small_q,reps,10)
experiment13 = ('experiment13',small_n,m_range,small_M,large_q,reps,10)
experiment14 = ('experiment14',small_n_range,m,small_M,small_q,reps,10)
# experiment15 = ('experiment15',range(2,11),m,small_M,small_q,reps,10)
### GENERATE FILES
if generate:
generate_experiment_files(*experiment8)
generate_experiment_files(*experiment9)
generate_experiment_files(*experiment10)
generate_experiment_files(*experiment11)
generate_experiment_files(*experiment12)
generate_experiment_files(*experiment13)
generate_experiment_files(*experiment14)
# generate_experiment_files(*experiment15)
### RUN EXPERIMENTS
if allocate:
# VARY q
run_allocate_test(experiment8, 4)
run_allocate_test(experiment9, 4)
# VARY n
run_allocate_test(experiment10, 1)
run_allocate_test(experiment11, 1)
# VARY m
run_allocate_test(experiment12, 2)
run_allocate_test(experiment13, 2)
run_allocate_test(experiment14, 1)
# run_allocate_test(experiment15, 1)
### PLOT GRAPHS
if plot:
import matplotlib.pyplot as plt
# plot experiments 8 and 9
for i in [8, 9]:
exp_name = 'experiment{}'.format(i)
f = 'allocate'
x_coords, y_coords = get_data(exp_name, f, 0, 1)
fig = plt.figure(figsize=(7, 4))
plt.clf()
plt.scatter(x_coords, y_coords, marker="+")
plt.ylabel('time (in seconds)')
plt.xlabel('avg number (B) of bids per bidder')
plt.xticks(range(0,501,100), range(0, 1251, 250))
plt.xlim(0, 520)
# plt.title("Experiment {}: Allocation".format(i))
filename = 'experiments/figures/{}-{}.eps'.format(exp_name, f)
plt.savefig(filename, format='eps')
# plot experiments 10 and 11
for i in [10, 11]:
exp_name = 'experiment{}'.format(i)
f = 'allocate'
x_coords, y_coords = get_data(exp_name, f, 0, 1)
fig = plt.figure(figsize=(7, 4))
plt.clf()
plt.scatter(x_coords, y_coords, marker="+")
plt.ylabel('time (in seconds)')
plt.xlabel('number of goods (n)')
plt.xticks(range(0,51,5))
plt.xlim(2.5, 52.5)
# plt.title("Experiment {}: Allocation".format(i))
filename = 'experiments/figures/{}-{}.eps'.format(exp_name, f)
plt.savefig(filename, format='eps')
# plot experiments 12 and 13
for i in [12, 13]:
exp_name = 'experiment{}'.format(i)
f = 'allocate'
x_coords, y_coords = get_data(exp_name, f, 0, 1)
fig = plt.figure(figsize=(7, 4))
plt.clf()
plt.scatter(x_coords, y_coords, marker="+")
plt.ylabel('time (in seconds)')
plt.xlabel('number of bidders (m)')
plt.xticks(range(0,21,2))
plt.xlim(1, 21)
# plt.title("Experiment {}: Allocation".format(i))
filename = 'experiments/figures/{}-{}.eps'.format(exp_name, f)
plt.savefig(filename, format='eps')
# plot experiments 14 (for 15 change number)
for i in [14]:
exp_name = 'experiment{}'.format(i)
f = 'allocate'
x_coords, y_coords = get_data(exp_name, f, 0, 1)
x_coords, y2_coords = get_data("experiment15", f, 0, 1)
fig = plt.figure(figsize=(7, 4))
plt.clf()
plt.scatter(x_coords, y_coords[:len(x_coords)], marker="+")
plt.scatter(x_coords, y2_coords, marker="x")
plt.ylabel('time (in seconds)')
plt.xlabel('number of goods (n)')
plt.xticks(range(1,11))
# plt.title("Experiment {}: Allocation".format(i))
filename = 'experiments/figures/{}-{}.eps'.format(exp_name, f)
plt.savefig(filename, format='eps')