checkpoint frequency is "savcrd_freq" ?

[nyubachi]

Hello,

I salute you all for your excellent efforts.

On a very minor point, for example, in the file "Protein_ligand.ipynb", in "Runs a Production MD simulation (NPT ensemble) after equilibration", the following code

print("\nSave coordinates each " + str(savcrd_freq))
print("\tSave checkpoint each " + str(savcrd_freq))
print("Print to log file " + str(print_freq))

I was a little concerned that the checkpoint frequency is "savcrd_freq". This may need to be corrected, as I feel that checkpoints are actually created for each stride.

Best regards,

nyubachi

[pablo-arantes]

Hi @nyubachi,

Sorry for the delay. I've reviewed the code and confirmed that the checkpoint is being saved correctly.

    ##################################
    # Writing last frame information of stride
    print("\n> Writing state file (" + str(rst_file) + ")...")
    state = simulation.context.getState( getPositions=True, getVelocities=True )
    with open(rst_file, 'w') as f:
        f.write(XmlSerializer.serialize(state))

    last_frame = int(nsteps/nsavcrd)
    print("> Writing coordinate file (" + str(pdb_file) + ", frame = " + str(last_frame) + ")...")
    positions = simulation.context.getState(getPositions=True).getPositions()
    PDBFile.writeFile(simulation.topology, positions, open(pdb_file, 'w'))

print("\n> Finished!\n")

The main correction was made to the log statements to ensure the right message about file saving is provided to the user. I've applied the changes to the file

Thank you.