Load model from txt file

[Alberto Manzini]

Hi everyone, I've saved my model in a txt file using method

model.ExportToFile(filename)

I don't understand how to load back my model from this file. I assume I have to load the proto message from file and setting it to a (empty) model, but I'm looking for the exact python commands to follow

Thanks for any help

[Laurent Perron]

There are no api. This is used mostly for debugging purposes.

--
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/583adb61-1a9a-449f-ab56-44265d58454cn%40googlegroups.com.

[watchdogs132]

from ortools.sat.python import cp_model
from google.protobuf import text_format

model = cp_model.CpModel()
with open("model.txt", "r") as file:
    text_format.Parse(file.read(), model.Proto())

[Alberto Manzini]

Cool, many thanks!

[Laurent Perron]

The question is to recover solutions.

You can use model.Get{Int|Bool|Interval}VarFromProtoIndex to rebuild variables to use to recover solutions.

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

To view this discussion on the web visit https://groups.google.com/d/msgid/or-tools-discuss/eb4391f2-37b6-4ba4-a43f-066275d68eb4n%40googlegroups.com.

[Alberto Manzini]

Hi, I was looking for a method to export and load the model, not the solutions, to debug it

Thank you both

[J. E. Marca]

I tried this advice and it didn't work for me.  

My solution is this:

```python

    from ortools.sat.python import cp_model_helper as cmh, cp_model

    test_file = "tests/data/dumpmodel_20260401_20260402_CpSolverStatus.INFEASIBLE.pbtxt"
    model = cp_model.CpModel()
    model_proto: cmh.CpModelProto = model.Proto()

    with open(test_file, "r") as file:
        model_proto.parse_text_format(str(file.read()))

        model_constraints = model_proto.constraints
        model_variables = model_proto.variables
        print(f"have {len(model_constraints)}  constraints")
        assert len(model_constraints) > 1
        print(f"have {len(model_variables)}  variables")
        assert len(model_variables) > 1
```

[Laurent Perron]

def clone(self) -> "CpModel":

"""Reset the model, and creates a new one from a CpModelProto instance."""

clone = CpModel()

clone.proto.copy_from(self.proto)

clone.rebuild_constant_map()

return clone

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

--
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 visit https://groups.google.com/d/msgid/or-tools-discuss/c59cf392-28d2-49a7-9e42-3d856ff7afb4n%40googlegroups.com.

[J. E. Marca]

Ah yes, resetting the model is probably important if you want to actually use the model in a solver run.

In my case, I am not interested in re-running the model (it is known to be infeasible).  Instead, I am writing code to inspect it in order to offer the user more details on what is causing the infeasibility.

I just needed to load the pbtxt file in my testing regime order to run tests on my various constraint and variable inspection methods.

Regards,

James