Run Surface Evolver via signac-flow

[adowling2]

Steps:

• Run surface evolver using command line
• Review steps
• Look at documentation for signac-flow commands
• Run signac flow for writing SE geometry
• Run signac in a linux os
• Add a signac operation to run SE using command line
• Add a signac operation to extract features from output files of SE simulation.

[adowling2]

@nil23121994 You should get an email with this message.

[adowling2]

Email from @nil23121994:

Hello Dr. Dowling,

I hope you are doing good. I have tried to integrate signac-flow with the geometry writer function in the codes attached below as a zip file. I tried to push it using github desktop but I was getting the error whose screenshot I have attached in this mail as "giterror".. Is this an error because we moved the repository to MSELab. I tried to look for that repository in our labpage but was not able to access it.

The zip folder attached contains the geometry writer function file and its dependencies. Apart from it following are the scripts that were used for integration with signac

1. init.py - initializing the signac workspace
2. project.py - Defines the operations to be executed during running jobs
3. main.ipynb - I call these scripts in a main jupyter notebook file.

I ran these codes but I am getting an error now which I am unable to resolve. The code writes the geometry file but it doesn't write it in the desired location but simply puts them in the working directory. I am attaching the error returned as "signacerror" image file.

I have also found a way to run the evolver dat file using a commandline and here is the syntax for that.

evolver [-adehimqwxy] [-f file] [-pN] [datafile]

(https://manpages.ubuntu.com/manpages/precise/en/man1/evolver.1.html)

[adowling2]

Email from @nil23121994:
hese are the steps I used for running signac flow

1. init.py - I defined a file init.py that is used to initialize the workspace. In this step I define the different statepoints and initialize the jobs in a loop. Once I run this step different folders are created and they contain a json file containing the information about the setpoints.
2. project.py - Following initialization I created a script project.py that contains the operations that I need to arry out using the setpoints. In thse operations I create a function that takes in the job as argument (statepoints) and writes the surface evolver initialization file.

I first defined the directory where I want to keep my data. I then ran the init.py file. I next ran the project.py file. This is the error I get while running the files

signacerror.JPG

I ran both these operations in the anaconda prompt and also ran them using a main jupyter notebook Do you think, I should run these steps on an ubuntu based interface instead of a windows system?

[adowling2]

[NilaySystemsBio]

1. The files init.py and project.py are in the folder Tissue Cartography/image-cellular-modeling/project parameterScreening/tissue-cartography-project.
2. First, I ranpython init.py in the Anaconda prompt.
3. Next, I ran python project.py run in the Anaconda prompt.
4. Finally, I ran python project.py submit in the Anaconda prompt.
5. We checked the status next using python project.py status -d. Below is the output.

Status:

Based on the tutorial, we were expecting a letter in the [ ] in the operation column.

@adowling2

[adowling2]

@bbefort @rsdefever Any ideas? Perhaps this is a quick fix.

[bbefort]

My screen looks like this if all of my available operations are completed. When you run python project.py status, do you still have available operations? If so, then it is probably an issue with an operation. Do you have project decorators that will indicate when a specific operation is complete? If you add these, the labels column will show an operation is complete. Also, if you submit anything on the crc, you should get a .o file. What does this say?

If you don't still have available operations, maybe they were all completed when you ran python project.py run?
If you don't specify any specific operation in python project.py run, I think it will run all available operations on the crc front end. Then, if you submit, nothing will need to run...

@rsdefever might have some other thoughts.

[rsdefever]

@nil23121994 What is the full traceback from here?

The files init.py and project.py are in the folder Tissue Cartography/image-cellular-modeling/project parameterScreening/tissue-cartography-project.

As an FYI you can link directly to files in a GH issue :)

Looking this line:

def SE_file_exist_check(job):
	return os.path.isfile("wingDisc.fe")

needs to be defined as a label, like

@FlowProject.label
def SE_file_exist_check(job):
	return os.path.isfile("wingDisc.fe")

I also normally subclass FlowProject to Project (or whatever name you like). See an example here.

class Project(FlowProject):
    pass

Those are the first issues I see. One other question. How are you trying to run the flow operation?

Edit: I see something about windows. Are you running this locally or on CRC. If local, what OS/how are you running it?

[NilaySystemsBio]

@bbefort I do not have any project decorators in my project.py file currently. I just have added a check to see if the operation has been completed.
@rsdefever I am running the code locally in a windows 7 os using anaconda.