docplex的安装与学习使用

 docplex的安装

我是在Anaconda和Python3.10环境上安装的

conda create -p xxx/cplex python=3.10
pip install cplex -i https://pypi.doubanio.com/simple
pip install docplex -i https://pypi.doubanio.com/simple

 docplex的初级使用：

代码来自(181条消息) anaconda环境下Cplex及Docplex安装流程-CSDN博客

# 导入库
import time
import random
import docplex.mp.model as cpx
import pandas as pd


start_time = time.time()

# 创建模型
model = cpx.Model()
# 创建变量列表
X = model.continuous_var_list([i for i in range(0, 2)], lb=0, name='X')
# 设定目标函数
model.minimize(2 * X[0] + 3 * X[1])
# 添加约束条件
model.add_constraint(3 * X[0] + X[1] >= 30)
model.add_constraint(X[0] - X[1] <= 10)
model.add_constraint(X[1] >= 1)
# 求解模型
sol = model.solve()
# 打印结果
print(sol)

end_time = time.time()
print(round(end_time - start_time, 4))

目录

 docplex的安装

 docplex的初级使用：

docplex的高级建模：

参考：

---

‘’代码主要来自python运筹优化（一）：Cplex for python使用简介 - 知乎 (zhihu.com)

# 导入库
import time
import random
import docplex.mp.model as cpx
import pandas as pd

start_time = time.time()

n = 10
m = 5
n1 = 6

set_I = range(1, n + 1)
set_J = range(1, m + 1)

c = {(i, j): random.normalvariate(0, 1) for i in set_I for j in set_J}
a = {(i, j): random.normalvariate(0, 5) for i in set_I for j in set_J}
l = {(i, j): random.randint(0, 10) for i in set_I for j in set_J}
u = {(i, j): random.randint(10, 20) for i in set_I for j in set_J}
b = {j: random.randint(0, 30) for j in set_J}

# 创建模型
opt_model = cpx.Model(name="MIP Model")
# 创建变量-way-1
# if x is Continuous
x_vars = {(i, j): opt_model.continuous_var(lb=l[i, j], ub=u[i, j], name="x_{0}_{1}".format(i, j)) for i in set_I for j
          in set_J}
# print(x_vars)
# 创建变量-way-2  一次性添加一组变量
# xb = {(i, j) for i in set_I for j in set_J} # 建立起下标的集合
# x_vars = opt_model.continuous_var_dict(xb, lb=l, ub=u, name="x")
# print(x_vars)
# 添加约束条件
constraints = {j: opt_model.add_constraint
(ct=opt_model.sum(a[i, j] * x_vars[i, j] for i in set_I) <= b[j],
 ctname="constraint_{0}".format(j))
               for j in set_J}
# 设定目标函数
objective = opt_model.sum(x_vars[i, j] * c[i, j]
                          for i in set_I
                          for j in set_J)
opt_model.minimize(objective)

# 求解模型:solving with local cplex
sol = opt_model.solve()
# # 打印结果
# print(sol)
# a = sol.to_string()


if sol is not None:
    opt_df = pd.DataFrame.from_dict(x_vars, orient="index", columns=["variable_object"])
    opt_df.index = pd.MultiIndex.from_tuples(opt_df.index, names=["column_i", "column_j"])
    opt_df.reset_index(inplace=True)

    # CPLEX
    opt_df["solution_value"] = opt_df["variable_object"].apply(lambda item: item.solution_value)
    opt_df.drop(columns=["variable_object"], inplace=True)
    opt_df.to_csv("./optimization_solution.csv")
else:
    print("solve failed")

end_time = time.time()
print(round(end_time - start_time, 4))

"""
防止约束条件是随机生成的,不存在满足约束的解,导致报错Model<MIP Model> did not solve successfully，
解决方法： 写到了while里面，当输出不为None时， break
"""
# 导入库
import time
import random
import docplex.mp.model as cpx
import pandas as pd

start_time = time.time()

sol = None

while True:
    n = 10
    m = 5
    n1 = 6

    set_I = range(1, n + 1)
    set_J = range(1, m + 1)

    c = {(i, j): random.normalvariate(0, 1) for i in set_I for j in set_J}
    a = {(i, j): random.normalvariate(0, 5) for i in set_I for j in set_J}
    l = {(i, j): random.randint(0, 10) for i in set_I for j in set_J}
    u = {(i, j): random.randint(10, 20) for i in set_I for j in set_J}
    b = {j: random.randint(0, 30) for j in set_J}
    # 创建模型
    opt_model = cpx.Model(name="MIP Model")
    # 创建变量
    # if x is Continuous
    x_vars = {(i, j): opt_model.continuous_var(lb=l[i, j], ub=u[i, j], name="x_{0}_{1}".format(i, j)) for i in set_I for
              j in set_J}
    # print(x_vars)
    # # 建立起下标的集合
    # xb = {(i, j) for i in set_I for j in set_J}
    # y_vars = opt_model.continuous_var_dict(xb, lb=l, ub=u, name="y")
    # print(y_vars)
    # 添加约束条件
    constraints = {j: opt_model.add_constraint
    (ct=opt_model.sum(a[i, j] * x_vars[i, j] for i in set_I) <= b[j],
     ctname="constraint_{0}".format(j))
                   for j in set_J}
    # 设定目标函数
    objective = opt_model.sum(x_vars[i, j] * c[i, j]
                              for i in set_I
                              for j in set_J)
    opt_model.minimize(objective)
    # 求解模型:solving with local cplex
    sol = opt_model.solve()
    if sol is not None:
        break
# # 打印结果
# print(sol)
# a = sol.to_string()

# if sol is not None:
opt_df = pd.DataFrame.from_dict(x_vars, orient="index", columns=["variable_object"])
opt_df.index = pd.MultiIndex.from_tuples(opt_df.index, names=["column_i", "column_j"])
opt_df.reset_index(inplace=True)

# CPLEX
opt_df["solution_value"] = opt_df["variable_object"].apply(lambda item: item.solution_value)
opt_df.drop(columns=["variable_object"], inplace=True)
opt_df.to_csv("./optimizastion_solution.csv")


end_time = time.time()
print(round(end_time - start_time, 4))

参考：

DOcplex介绍及安装（1） - 知乎 (zhihu.com)

(181条消息) anaconda环境下Cplex及Docplex安装流程-CSDN博客

python运筹优化（一）：Cplex for python使用简介 - 知乎 (zhihu.com)