Re: [or-tools-discuss] AttributeError: 'list' object has no attribute 'GetIntegerVarValueMap'

[Laurent Perron]

Too vague.

I cannot run your code as the data files are not included, and you do not specify the line where it fails.

Most likely you are passing a list where the solver expects an expression (x + y).

About speed, there is not much we can do. There is some pandas support, not sure it is fully optimized.

Laurent Perron | Operations Research | lpe...@google.com | (33) 1 42 68 53 00

Le mer. 13 déc. 2023 à 13:16, Julius Günther <juli...@outlook.de> a écrit :

I try to add batch sizes to the single-machine scheduling model.
I get the AttributeError: 'list' object has no attribute 'GetIntegerVarValueMap'.
What does that mean and how can I fix that?

Except for that, do you have suggestions on how to add variable Batch-sizes efficiently?

import pandas as pd

import openpyxl

from typing import Sequence

import collections

from absl import app

from absl import flags

from google.protobuf import text_format

from ortools.sat.python import cp_model

# Einlesen der Daten ---------------------------------------------------------------------------------------

excel_path = "C:\\Users\\juliu\\Nextcloud\\0_DPL_Guenther_Julius\\Praxisphase\\Eingabedatei.xlsx"

excel_path_2 = "C:\\Users\\juliu\\Nextcloud\\0_DPL_Guenther_Julius\\Praxisphase\\Ausgabedatei.xlsx"

sheet_name_import = "Daten uc4"

df1 = pd.read_excel(excel_path, sheet_name_import, header = None, usecols = "C:J", skiprows = 14, nrows = 8)

df2 = pd.read_excel(excel_path, sheet_name_import, header = None)

weekly_demand = df2.iloc[2, 2:10].astype(str).replace(("'"),("")).astype(int).tolist()

min_batch_size = df2.iloc[4, 2:10].astype(str).replace(("'"),("")).astype(int).tolist()

cycle_times = df2.iloc[10, 2:10].astype(str).replace(("'"),("")).astype(int).tolist()

setup_times = [[30]*len(weekly_demand)] # Start-Job (Dummy)

setup_times_no_dummy = df1.values.astype(int).tolist()

for element in setup_times_no_dummy:

    setup_times.append(element)

due_dates = df2.iloc[8, 2:10].astype(str).replace(("'"),("")).astype(int).tolist()

due_dates = [element *8*60 for element in due_dates]

release_dates = df2.iloc[9, 2:10].astype(str).replace(("'"),("")).astype(int).tolist()

release_dates = [element *8*60 for element in release_dates]

priorities = df2.iloc[11, 2:10].astype(str).tolist()

print(setup_times)

print(weekly_demand)

print(min_batch_size)

print(due_dates)

print(release_dates)

print(priorities)

# Verarbeiten der Daten-------------------------------------------------------------------------

min_durations = [a // b for a, b in zip(min_batch_size, cycle_times)]

max_durations = [a // b for a, b in zip(weekly_demand, cycle_times)]

num_jobs = [a // b for a, b in zip(weekly_demand, min_batch_size)]

job_durations = [[a] * b for a, b in zip(min_durations, num_jobs)]

num_jobs = len(job_durations)

all_jobs = range(num_jobs)

# Erstellen von Befehlszeilenargumenten -----------------------------------------------

_OUTPUT_PROTO = flags.DEFINE_string(

    "output_proto", "", "Output file to write the cp_model proto to."

)

_PARAMS = flags.DEFINE_string(

    "params",

    "log_search_progress:true",

    "Sat solver parameters.",

)

_PREPROCESS = flags.DEFINE_bool(

    "--preprocess_times", True, "Preprocess setup times and durations"

)

# Konfiguration der Lösungsausgabe ----------------------------------------------------

class SolutionPrinter(cp_model.CpSolverSolutionCallback):

    def __init__(self):

        cp_model.CpSolverSolutionCallback.__init__(self)

        self.__solution_count = 0

       

    def on_solution_callback(self):

        print(

            "Solution %i, time = %f s, objective = %i"

            % (self.__solution_count, self.WallTime(), self.ObjectiveValue())

        )

               

        self.__solution_count += 1

# Erstellen von Lösungsfunktion -----------------------------------------------------

def single_machine_scheduling():

   

    parameters = _PARAMS.value

    output_proto_file = _OUTPUT_PROTO.value

   

    # Erstellen des Lösungsmodells -------------------------------------------------

    model = cp_model.CpModel()

   

    # Festlegen des Horizonts ------------------------------------------------------

    horizon = sum(max_durations) + sum(

        max(setup_times[i][j] for i in range(num_jobs + 1)) for j in range(num_jobs)

    )

   

    print("Greedy horizon =", horizon)

    intervals = []

    starts = []

    ends = []

    # Erstellen der Variablen für jeden Job ----------------------------------------

    for job_id in all_jobs:

        product_start = []

        product_end = []

        for batch in range(max(len(zeile) for zeile in job_durations)):

            duration = job_durations[job_id][batch]

            release_date = release_dates[job_id]

            due_date = due_dates[job_id]

            print(

                "job %2i: start = %5i, duration = %4i, end = %6i"

                % (job_id, release_date, duration, due_date)

            )

            name_suffix = "_%i_%i" % (job_id, batch)

            start = model.NewIntVar(release_date, due_date, "s" + name_suffix)

            end = model.NewIntVar(release_date, due_date, "e" + name_suffix)

            interval = model.NewIntervalVar(start, duration, end, "i" + name_suffix)

            product_start.append(start)

            product_end.append(end)

            intervals.append(interval)

        starts.append(product_start)

        ends.append(product_end)

       

    print(starts)

    # Erstellen von Überlappungs-Constraint -----------------------------------------

    model.AddNoOverlap(intervals)

    # Erstellen von Verbindungen aller Jobs der Rüstmatrix --------------------------

    arcs = []

    for i in all_jobs:

        min_start_time = max(release_dates[i], setup_times[0][i])

        for batch in range(max(len(zeile) for zeile in job_durations)):

            start_lit = model.NewBoolVar("")

            arcs.append((0, batch + 1, start_lit))

            model.Add(starts[i][batch] == min_start_time).OnlyEnforceIf(start_lit)

            arcs.append((batch + 1, 0, model.NewBoolVar("")))

           

            for j in all_jobs:

                for batch2 in range(max(len(zeile) for zeile in job_durations)):

                    if i == j & batch == batch2:

                        continue

                    lit = model.NewBoolVar("%i_%i follows %i_%i" % (j, batch, i, batch2))

                    arcs.append((batch + 1, batch2 + 1, lit))

           

                    model.Add(starts[j][batch2] >= ends[i][batch] + setup_times[i + 1][j]).OnlyEnforceIf(lit)

                   

    # Erstellen von Verbindungs-Constraint -----------------------------------------

    model.AddCircuit(arcs)

    # Erstellen der Zeilfunktion ---------------------------------------------------

    makespan = model.NewIntVar(0, horizon, "makespan")

    model.AddMaxEquality(makespan, ends)

    model.Minimize(makespan)

    if output_proto_file:

        print("Writing proto to %s" % output_proto_file)

        with open(output_proto_file, "w") as text_file:

            text_file.write(str(model))

    # Erstellen des Solvers --------------------------------------------------------

    solver = cp_model.CpSolver()

    if parameters:

        text_format.Parse(parameters, solver.parameters)

       

    solution_printer = SolutionPrinter()

    # Übergabe des Modells an den Solver -------------------------------------------

    solver.Solve(model, solution_printer)

   

    for job_id in all_jobs:

        for batch in range(max(len(zeile) for zeile in job_durations)):

            print(

                "job %i_%i starts at %i end ends at %i"

                % (job_id + 1, batch + 1, solver.Value(starts[job_id][batch]), solver.Value(ends[job_id][batch]))

            )

       

# Erstellen der Hauptfunktion -----------------------------------------------------

def main(argv: Sequence[str]) -> None:

    if len(argv) > 1:

        raise app.UsageError("Too many command-line arguments.")

   

    single_machine_scheduling()

if __name__ == "__main__":

    app.run(main)

--
You received this message because you are subscribed to the Google Groups "or-tools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to or-tools-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/or-tools-discuss/a0421290-4bf3-4304-8b4a-e3d871876279n%40googlegroups.com.

[Laurent Perron]

No, which line in your code. My case works, it just receive a wrong argument

Le mer. 13 déc. 2023, 14:05, Julius Günther <juli...@outlook.de> a écrit :

I'm Sorry, here is the code with the data files.
The code fails in the cp_model file:

  File "C:\Users\juliu\AppData\Roaming\Python\Python311\site-packages\ortools\sat\python\cp_model.py", line 2326, in ParseLinearExpression
    coeffs_map, constant = cast(LinearExpr, linear_expr).GetIntegerVarValueMap()
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

AttributeError: 'list' object has no attribute 'GetIntegerVarValueMap'

import pandas as pd

import openpyxl

from typing import Sequence

import collections

from absl import app

from absl import flags

from google.protobuf import text_format

from ortools.sat.python import cp_model

weekly_demand = [1250000, 1500000, 1250000, 2000000, 1000000, 750000, 2000000, 750000]

min_batch_size = [250000, 300000, 250000, 400000, 200000, 150000, 400000, 150000]

cycle_times = [10000, 10000, 10000, 10000, 10000, 10000, 10000, 10000]

setup_times = [[30, 30, 30, 30, 30, 30, 30, 30],

               [0, 45, 30, 35, 40, 25, 30, 35],

               [40, 0, 30, 45, 20, 50, 25, 30],

               [30, 35, 0, 20, 30, 40, 30, 30],

               [30, 25, 30, 0, 35, 25, 35, 40],

               [25, 20, 35, 30, 0, 40, 45, 35],

               [35, 25, 20, 35, 30, 0, 40, 35],

               [30, 20, 15, 20, 30, 40, 0, 25],

               [50, 25, 30, 25, 20, 35, 30, 0]]

due_dates = [2400, 2400, 2400, 2400, 2400, 2400, 2400, 2400]

release_dates = [0, 0, 0, 0, 0, 0, 0, 0]

To view this discussion on the web visit https://groups.google.com/d/msgid/or-tools-discuss/f6dcc04f-6ede-4ba5-a634-081893e8ea6fn%40googlegroups.com.

[Laurent Perron]

of course :-(

Laurent Perron | Operations Research | lpe...@google.com | (33) 1 42 68 53 00

Le mer. 13 déc. 2023 à 14:17, Julius Günther <juli...@outlook.de> a écrit :

I guess it's in this line:     model.AddMaxEquality(makespan, ends)

I think because ends is a 2D List?

To view this discussion on the web visit https://groups.google.com/d/msgid/or-tools-discuss/4250287d-8fe6-4920-9ebc-6556e29b60d3n%40googlegroups.com.

[Laurent Perron]

  model.Add(starts[i][batch] == min_start_time).OnlyEnforceIf(start_lit)

Risky: are you sure no job will still be active on that machine at min_start_time ? I would use >=

For batches, you do not impose an order ? batch1 before batch2...

Le mer. 13 déc. 2023 à 15:11, Julius Günther <juli...@outlook.de> a écrit :

Another Question:
When I model the arcs, I still connect Jobs with themselves: Circuit/Route constraint contains multiple self-loop involving node 1

I thought I fixed it with the if function?

    arcs = []

    for i in all_jobs:

        min_start_time = setup_times[0][i]

        for batch in range(max(len(zeile) for zeile in job_durations)):

            start_lit = model.NewBoolVar("")

            arcs.append((0, batch + 1, start_lit))

            model.Add(starts[i][batch] == min_start_time).OnlyEnforceIf(start_lit)

            arcs.append((batch + 1, 0, model.NewBoolVar("")))

           

            for j in all_jobs:

                for batch2 in range(max(len(zeile) for zeile in job_durations)):

                    if i == j and batch == batch2:

                        continue

                    lit = model.NewBoolVar("%i_%i follows %i_%i" % (j, batch, i, batch2))

                    arcs.append((batch + 1, batch2 + 1, lit))

           

                    model.Add(starts[j][batch2] >= ends[i][batch] + setup_times[i + 1][j]).OnlyEnforceIf(lit)

Julius Günther schrieb am Mittwoch, 13. Dezember 2023 um 14:27:23 UTC+1:

Sorry my fault. 
Code still doesn't work, but thanks.

To view this discussion on the web visit https://groups.google.com/d/msgid/or-tools-discuss/64b2f295-13f0-42d6-9871-cd77b7eb1889n%40googlegroups.com.

--

--Laurent