Skip to content

bug in Periodic.py? #34

Description

@mkrompiec

I'm trying to run an MD simulation in PBC, using ChemSpider-trained network:

def EnAndForce(z_, x_, nreal_, DoForce = True):
mtmp = Mol(z_,x_)
en,f = manager.EvalBPDirectEEUpdateSinglePeriodic(mtmp, PARAMS["AN1_r_Rc"],
PARAMS["AN1_a_Rc"], PARAMS["EECutoffOff"], nreal_,True)
return en[0], f[0]

lat=Lattice(np.eye(3)*70) # 70 angstrom box
m.properties["Lattice"]=lat
PARAMS["MDThermostat"] = "Nose"
PARAMS["MDTemp"] = 300
PARAMS["MDdt"] = 0.1 # fs
PARAMS["RemoveInvariant"]=True
PARAMS["MDV0"] = "Random"
PARAMS["MDMaxStep"] = 10000
PF = PeriodicForce(m,np.eye(3)*70) #m.properties["Lattice"])
PF.BindForce(EnAndForce, 70.0)
PARAMS["MDTemp"] = 300.0
PARAMS["MDdt"] = 0.05 # In fs.
traj = PeriodicVelocityVerlet(PF,"sim_300K")
traj.Prop()

Traceback (most recent call last):
File "runmd.py", line 79, in
traj.Prop()
File "/mainfs/scratch/mk1a18/TensorMol/TensorMol/Simulations/PeriodicMD.py", line 144, in Prop
LOGGER.info("Step: %i time: %.1f(fs) (kJ/mol): %.5f <|a|>(m/s2): %.5f (Eh): %.5f (kJ/mol): %.5f Rho(g/cm**3): %.5f Teff(K): %.5f", step, self.t, self.KE/1000.0, np.linalg.norm(self.a) , self.EPot, self.KE/1000.0+self.EPot*KJPERHARTREE,self.Density(), Teff)
File "/mainfs/scratch/mk1a18/TensorMol/TensorMol/Simulations/PeriodicMD.py", line 106, in Density
return self.PForce.Density()
File "/mainfs/scratch/mk1a18/TensorMol/TensorMol/ForceModifiers/Periodic.py", line 309, in Density
m = np.array(map(lambda x: ATOMICMASSES[x-1], self.mol0.atoms))*1000.0
TypeError: unsupported operand type(s) for *: 'map' and 'float'

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions