scf_theoretical/grid_search.py at main · lx249/scf_theoretical · GitHub

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
"""
Runing simulation by using different combinations of input parameters.
Author: Liming Xu
Email: lx249@cam.ac.uk
"""

# %%
import itertools
import numpy as np
from simulation import SCFSimulation
from utils import load_config_file


def simconfig_generator(input_params):
    """
    Enumerate input parameters for grid search.
    """
    # Part one - basics
    t_max                    = input_params["t_max"]
    lst_topology             = input_params["network_topology"]
    lst_homogeneous          = input_params["homogeneous"]
    lst_operation_fee        = input_params["operation_fee"]

    # Part two - demand generation
    lst_demand_distribution  = input_params["demand_distribution"]
    demand_distribution      = lst_demand_distribution[0]
    lst_demand_mean          = input_params["distribution_params"][demand_distribution]["mean"]
    lst_demand_sigma         = input_params["distribution_params"][demand_distribution]["sigma"]

    # Part three - company power and marker shares
    powers                   = input_params["powers"]
    lst_small_market_shares  = [ 1 ]
    lst_medium_market_shares = input_params["market_shares"]["medium"]
    lst_large_market_shares  = input_params["market_shares"]["large"]

    # Part four - financing
    lst_financed             = input_params["financed"]
    lst_loan_repayment_time  = input_params["loan_repayment_time"]
    lst_bank_annual_rate     = input_params["bank_annual_rate"]
    lst_invoice_annual_rate  = input_params["invoice_annual_rate"]
    lst_invoice_term         = input_params["invoice_term"]

    # Part five - financing threshold
    lst_paradigm             = input_params["paradigm"]
    lst_ma_window_size       = input_params["moving_average"]["window_size"]

    # Input parameter enumeration
    basics = itertools.product(
        lst_topology,
        lst_homogeneous,
        lst_operation_fee)

    demand_generation = itertools.product(
        lst_demand_distribution,
        lst_demand_mean,
        lst_demand_sigma)

    all_market_share_combs = itertools.product(
        lst_small_market_shares,
        lst_medium_market_shares,
        lst_large_market_shares)
    market_shares = [(s, m, l)
                     for s, m, l in all_market_share_combs
                     if s <= m and m <= l]

    financing_params = itertools.product(
        lst_loan_repayment_time,
        lst_bank_annual_rate,
        lst_invoice_annual_rate,
        lst_invoice_term
    )

    financing = [(False, 0, 0, 0, 0)]  # 0 as a placeholder
    for p1, p2, p3, p4 in financing_params:
        financing.append((True, p1, p2, p3, p4))

    financing_threshold = [("reactive", 1)]  # 1 as a placeholder
    for ws in lst_ma_window_size:
        financing_threshold.append(("proactive", ws))

    sim_configs = []
    sim_id = 0
    for v1, v2, v3, v4, v5 in itertools.product(
        basics,
        demand_generation,
        market_shares,
        tuple(financing),
        tuple(financing_threshold)
    ):
        sim_id += 1
        sim_config = {}
        sim_config["sim_id"] = sim_id
        sim_config["t_max"] = t_max
        sim_config["network_topology"] = v1[0]
        sim_config["homogeneous"] = v1[1]
        sim_config["operation_fee"] = v1[2]

        sim_config["demand_distribution"] = v2[0]
        sim_config["distribution_params"] = {"mean": v2[1], "sigma": v2[2]}

        sim_config["financed"] = v4[0]
        sim_config["loan_repayment_time"] = v4[1]
        sim_config["bank_annual_rate"] = v4[2]
        sim_config["invoice_annual_rate"] = v4[3]
        sim_config["invoice_term"] = v4[4]
        sim_config["paradigm"] = v5[0]
        sim_config["window_size"] = v5[1]
        sim_config["powers"] = powers
        sim_config["market_shares"] = v3

        sim_configs.append(sim_config)

    return sim_configs


def single_run_params(sim_config):
    """
    Generate the parameters for a signle simulation running.
    """
    sim_id              = sim_config["sim_id"]
    topology            = sim_config["network_topology"]
    homogeneous         = sim_config["homogeneous"]
    t_max               = sim_config["t_max"]
    financed            = sim_config["financed"]
    paradigm            = sim_config["paradigm"]
    operation_fee       = sim_config["operation_fee"]
    loan_repayment_time = sim_config["loan_repayment_time"]
    bank_annual_rate    = sim_config["bank_annual_rate"]
    invoice_annual_rate = sim_config["invoice_annual_rate"]
    invoice_term        = sim_config["invoice_term"]
    window_size         = sim_config["moving_average"]["window_size"]
    powers              = sim_config["powers"]
    market_shares       = sim_config["market_shares"]

    # Demand generator by normal distribution
    demand_distribution = sim_config["demand_distribution"]
    distribution_params = sim_config["distribution_params"][demand_distribution]

    params = {
        "t_max": t_max,
        "financed": financed,
        "paradigm": paradigm,
        "operation_fee": operation_fee,
        "loan_repayment_time": loan_repayment_time,
        "bank_annual_rate": bank_annual_rate,
        "invoice_annual_rate": invoice_annual_rate,
        "invoice_term": invoice_term,
        "window_size": window_size,
        "powers": powers,
        "market_shares": market_shares,
        "demand_distribution": demand_distribution,
        "distribution_params": distribution_params
    }
    return sim_id, topology, homogeneous, params


def grid_search_params(config_file):
    """
    Enumerate all valid combinations of parameters for grid search.
    """


# Network configuraitons
network_config = load_config_file("configs/network_config.yaml")

# # Single run
# sim_config = load_config_file("configs/simulation_config.yaml")
# (sim_id, topology, homogeneous, input_params) = single_run_params(sim_config)

# # New a simulation instance
# sim = SCFSimulation(sim_id=0,
#                     topology,
#                     homogeneous,
#                     network_config,
#                     **input_params)
# # Run the simulation
# sim.run()


# Grid search
input_params = load_config_file("configs/grid_search_inputs.yaml")
sim_configs = simconfig_generator(input_params)

for config in sim_configs:
    sim_id = config["sim_id"]
    topology = config["network_topology"]
    homogeneous = config["homogeneous"]

    config.pop("sim_id")
    config.pop("network_topology")
    config.pop("homogeneous")

    # Run simulation
    sim = SCFSimulation(sim_id,
                        topology,
                        homogeneous,
                        network_config,
                        **config)
    # Run the simulation
    sim.run()

    if sim_id == 2: break