Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Yes see this

"""This example show how to run an OPL model with its dat files, then query the post processing IloTupleSets.It also shows how to generate CSV files to be used with DO4DSX, the optimization addon to IBM DSXL"""from doopl.factory import *import osfrom os.path import dirname, abspath, joinDATADIR = join(dirname(abspath(__file__)), 'data')mod = join(DATADIR, "mulprod.mod")dat = join(DATADIR, "mulprod.dat")dat2 = join(DATADIR, "mulprod2.dat")gen_dir = os.path.join(DATADIR, "generated")if not os.path.isdir(gen_dir):    os.makedirs(gen_dir)with create_opl_model(model=mod, data=[dat, dat2]) as opl:    opl.run()    print(opl)    print(opl.objective_value)    print(opl.get_table("OutsideSolution"))    print(opl.get_table("InvSolution"))    print(opl.report)    print('Generating csv in %s' % os.path.join(gen_dir, "capacity.csv"))    opl.get_table("Capacity").to_csv(os.path.join(gen_dir, "capacity.csv"), index=False)

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Yes see this

"""This example show how to run an OPL model with its dat files, then query the post processing IloTupleSets.It also shows how to generate CSV files to be used with DO4DSX, the optimization addon to IBM DSXL"""from doopl.factory import *import osfrom os.path import dirname, abspath, joinDATADIR = join(dirname(abspath(__file__)), 'data')mod = join(DATADIR, "mulprod.mod")dat = join(DATADIR, "mulprod.dat")dat2 = join(DATADIR, "mulprod2.dat")gen_dir = os.path.join(DATADIR, "generated")if not os.path.isdir(gen_dir):    os.makedirs(gen_dir)with create_opl_model(model=mod, data=[dat, dat2]) as opl:    opl.run()    print(opl)    print(opl.objective_value)    print(opl.get_table("OutsideSolution"))    print(opl.get_table("InvSolution"))    print(opl.report)    print('Generating csv in %s' % os.path.join(gen_dir, "capacity.csv"))    opl.get_table("Capacity").to_csv(os.path.join(gen_dir, "capacity.csv"), index=False)

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Hi,

I tried to install doopl using pip command.

pip instal doopl

However, it fails giving the following messesage:

ERROR: Could not find a version that satisfies the requirement doopl (from versions: none)
ERROR: No matching distribution found for doopl

Is there any way to install?

Yes see this

"""This example show how to run an OPL model with its dat files, then query the post processing IloTupleSets.It also shows how to generate CSV files to be used with DO4DSX, the optimization addon to IBM DSXL"""from doopl.factory import *import osfrom os.path import dirname, abspath, joinDATADIR = join(dirname(abspath(__file__)), 'data')mod = join(DATADIR, "mulprod.mod")dat = join(DATADIR, "mulprod.dat")dat2 = join(DATADIR, "mulprod2.dat")gen_dir = os.path.join(DATADIR, "generated")if not os.path.isdir(gen_dir):    os.makedirs(gen_dir)with create_opl_model(model=mod, data=[dat, dat2]) as opl:    opl.run()    print(opl)    print(opl.objective_value)    print(opl.get_table("OutsideSolution"))    print(opl.get_table("InvSolution"))    print(opl.report)    print('Generating csv in %s' % os.path.join(gen_dir, "capacity.csv"))    opl.get_table("Capacity").to_csv(os.path.join(gen_dir, "capacity.csv"), index=False)

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

See

https://community.ibm.com/community/user/blogs/vincent-beraudier/2022/06/22/opl-connector-for-python-doopl-available-on-github?hlmlt=VT

Hi,

I tried to install doopl using pip command.

pip instal doopl

However, it fails giving the following messesage:

ERROR: Could not find a version that satisfies the requirement doopl (from versions: none)
ERROR: No matching distribution found for doopl

Is there any way to install?

Yes see this

"""This example show how to run an OPL model with its dat files, then query the post processing IloTupleSets.It also shows how to generate CSV files to be used with DO4DSX, the optimization addon to IBM DSXL"""from doopl.factory import *import osfrom os.path import dirname, abspath, joinDATADIR = join(dirname(abspath(__file__)), 'data')mod = join(DATADIR, "mulprod.mod")dat = join(DATADIR, "mulprod.dat")dat2 = join(DATADIR, "mulprod2.dat")gen_dir = os.path.join(DATADIR, "generated")if not os.path.isdir(gen_dir):    os.makedirs(gen_dir)with create_opl_model(model=mod, data=[dat, dat2]) as opl:    opl.run()    print(opl)    print(opl.objective_value)    print(opl.get_table("OutsideSolution"))    print(opl.get_table("InvSolution"))    print(opl.report)    print('Generating csv in %s' % os.path.join(gen_dir, "capacity.csv"))    opl.get_table("Capacity").to_csv(os.path.join(gen_dir, "capacity.csv"), index=False)

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?

For doopl, this example shows how to run a model with its .dat files : https://github.com/IBMDecisionOptimization/doopl/blob/master/src/samples/examples/Mulprod.py

Dear Alex

Is there any way to load both of .mod and .dat files?

Hi,

from python you can call oplrun but you can also rely on doopl

See this example

from doopl.factory import *# DataBuses=[        (40,500),        (30,400)        ]# Create an OPL model from a .mod filewith create_opl_model(model="zootupleset.mod") as opl:    # tuple can be a list of tuples, a pandas dataframe...    opl.set_input("buses", Buses)    # Generate the problem and solve it.    opl.run()    # Get the names of post processing tables    print("Table names are: "+ str(opl.output_table_names))    # Get all the post processing tables as dataframes.    for name, table in iteritems(opl.report):        print("Table : " + name)    for t in table.itertuples(index=False):            print(t)    # nicer display    for t in table.itertuples(index=False):        print(t[0]," buses ",t[1], "seats")

You can also import a model generated with opl in a docplex model : https://github.com/AlexFleischerParis/zoodocplex/blob/master/zooimportoplmodelandaddpythonobjectiveandconstraint.py

from docplex.mp.model import Modelfrom docplex.mp.model_reader import ModelReaderfrom doopl.factory import *# Create an OPL model from a .mod filewith create_opl_model(model="zoooplwithoutobjective.mod") as opl:    # Generate the problem and solve it.    opl.run()mdl = ModelReader.read_model('zoowithoutobj.lp', model_name='zoo')nbbus40 = mdl.find_matching_vars('nbBus40')[0]nbbus30 = mdl.find_matching_vars('nbBus30')[0]#add docplex constraint        mdl.add(nbbus40<=4)#add docplex objectivemdl.minimize(nbbus40*500 + nbbus30*400)msol=mdl.solve()print(msol[nbbus40]," buses 40 seats")print(msol[nbbus30]," buses 30 seats")

and you can also turn an integer decision variable to a float decision variable in docplex

from docplex.mp.model import Modelmdl = Model(name='buses')nbbus40 = mdl.integer_var(name='nbBus40')nbbus30 = mdl.integer_var(name='nbBus30')mdl.add_constraint(nbbus40*40 + nbbus30*30 >= 300, 'kids')mdl.minimize(nbbus40*500 + nbbus30*400)mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)nbbus30.set_vartype('C')print("relax nbbus30 to continuous")mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)for v in mdl.iter_continuous_vars():    print(v," = ",v.solution_value)print("set  nbbus30 back to integer")nbbus30.set_vartype('I')mdl.solve(log_output=False,)for v in mdl.iter_integer_vars():    print(v," = ",v.solution_value)

Dear Alex

Thank you for your reply.

My previous program is written in OPL Script and uses the ConvertAllIntVars() method for continuous relaxation, as in the second example you showed. However, OPL Script is not well-suited for integration with machine learning. I am planning to use the output of a random forest for some of the decisions, and Python is much more convenient for implementing this part. I also considered using docplex, but it would require translating the contents of the existing .modfile into Python, which is time-consuming.

Since performing continuous relaxation via the oplrun command seems difficult, directly editing the existing .mod file for continuous relaxation appears to be the easier approach.

Hi

in docplex you can relax integrity constraints as can be seen at https://github.com/AlexFleischerParis/zoodocplex/blob/master/zoorelaxintegrity.py

from docplex.mp.model import Modelfrom docplex.mp.relax_linear import LinearRelaxerdef make_bus_model():    mdl = Model(name='buses')    nbbus40 = mdl.integer_var(name='nbBus40')    nbbus30 = mdl.integer_var(name='nbBus30')    mdl.add_constraint(nbbus40 * 40 + nbbus30 * 30 >= 300, 'kids')    mdl.minimize(nbbus40 * 500 + nbbus30 * 400)    return mdlif __name__ == '__main__':    bm1 = make_bus_model()    bm1.print_information()    s1 = bm1.solve(log_output=True)    s1.display()    bmr = LinearRelaxer.make_relaxed_model(bm1)    bmr.print_information()    rs = bmr.solve(log_output=True)    rs.display()    duals = bmr.get_constraint_by_name("kids").dual_value    print("dual of the 300 kids constraint = ",duals)    for v in bmr.iter_continuous_vars():       print(v," = ",v.solution_value)       rc=v.reduced_cost       print("reduced cost =",rc)

You can also do that directly in OPL as can be seen at https://github.com/AlexFleischerParis/zooopl/blob/master/zoorelaxinteger.mod

    int nbKids=300;    float costBus40=500;    float costBus30=400;         dvar int+ nbBus40;    dvar int+ nbBus30;         minimize     costBus40*nbBus40  +nbBus30*costBus30;         subject to    {     ctKids:40*nbBus40+nbBus30*30>=nbKids;    }    main {      var status = 0;      thisOplModel.generate();      if (cplex.solve()) {        writeln("Integer Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                 }      // relax integrity constraint      thisOplModel.convertAllIntVars();           if (cplex.solve()) {        writeln("Relaxed Model");           writeln("OBJECTIVE: ",cplex.getObjValue());                  writeln("dual of the kids constraint = ",thisOplModel.ctKids.dual);        writeln("reduced costs for nbbus40 : ",thisOplModel.nbBus40.reducedCost);        writeln("reduced costs for nbbus30 : ",thisOplModel.nbBus30.reducedCost);      }                }

regards

Previously, I was using OPL Script, but I decided to switch to Python in order to incorporate machine learning. However, since I still want to use the existing OPL model, I am calling the oplrun command from Python using the subprocessmodule. The OPL model is an MIQP.

In the previous OPL Script, there was a part where the convertAllIntVars method was used to solve a continuous relaxation of the problem. I am having trouble figuring out how to solve the continuous relaxation using the oplruncommand. Do I have no choice but to manually convert the model into its relaxed form?