DO NOT MERGE: Workflow/firetasks for Equation of state thermal (request for review)

mpworks/drones/README.md

@@ -0,0 +1,9 @@
+# Drones package
+
+ The MPVaspDrone is currently used by the production workflow.
+
+The drones package is an extension of the pymatgen-db drone, which converts a VASP directory into a database dictionary. The MPVaspDrone adds a "post_process" method and modifies some of the default drone behavior. It would be better if this could extend the existing drone rather than repeat a lot of the pymatgen-db drone, but it was not workable at the time of its creation.
+
+For example, the signal detectors help tag extra things that have might gone wrong with the run, and put it in the key analysis.signals and analysis.critical_signals.
+
+Another thing the custom drone does is SNL management. In particular, for structure optimizations it adds a new SNL to the SNL database (the newly optimized structure). For static runs (where the structure doesn't change), a new SNL is not added. The packages also add keys like "snlgroup_changed" which check whether the new and old SNL match after the relaxation run.

mpworks/drones/mp_vaspdrone.py

@@ -21,7 +21,7 @@
 from pymatgen.entries.compatibility import MaterialsProjectCompatibility
 from pymatgen.entries.computed_entries import ComputedEntry
 from pymatgen.matproj.snl import StructureNL
-from pymatgen.io.vaspio.vasp_output import Vasprun, Outcar
+from pymatgen.io.vasp.outputs import Vasprun, Outcar
 from pymatgen.analysis.structure_analyzer import oxide_type
 
 
@@ -245,7 +245,13 @@ def process_fw(self, dir_name, d):
             d['snlgroup_id'] = fw_dict['spec']['snlgroup_id']
             d['vaspinputset_name'] = fw_dict['spec'].get('vaspinputset_name')
             d['task_type'] = fw_dict['spec']['task_type']
-
+            # Process data for deformed structures
+            if 'deformed' in d['task_type']:
+                d['deformation_matrix'] = fw_dict['spec']['deformation_matrix']
+                d['original_task_id'] = fw_dict['spec']['original_task_id']
+            if 'modified volume' in d['task_type']:
+                d['strainfactor'] = fw_dict['spec']['strainfactor']
+                d['original_task_id'] = fw_dict['spec']['original_task_id']
             if not self.update_duplicates:
                 if 'optimize structure' in d['task_type'] and 'output' in d:
                     # create a new SNL based on optimized structure

mpworks/dupefinders/README.md

@@ -0,0 +1,3 @@
+# Dupefinders package
+
+ The Dupefinders are currently used by the production workflow. They are used to implement the duplicate checking within the production FireWorks.

mpworks/examples/README.md

@@ -0,0 +1,3 @@
+# Examples package
+
+ These are meant to be more simple examples of workflows than the production MP Workflows. They are referred to in the MPWorks docs. These are only examples meant as learning tools and are not used in production. The "fireworks-vasp" repo is a different attempt at this.

mpworks/examples/firetasks_ex.py

@@ -26,9 +26,9 @@ def __init__(self, parameters):
         parameters = parameters if parameters else {}
         self.update(parameters)
         # get VaspJob objects from 'jobs' parameter in Firework
-        self.jobs = map(VaspJob.from_dict, parameters['jobs'])
+        self.jobs = parameters['jobs']
         # get VaspHandler objects from 'handlers' parameter in Firework
-        self.handlers = map(MontyDecoder().process_decoded, parameters['handlers'])
+        self.handlers = parameters['handlers']
         self.max_errors = parameters['max_errors']
 
     def run_task(self, fw_spec):

mpworks/examples/wf_ex.py

@@ -9,7 +9,7 @@
 from mpworks.firetasks.vasp_setup_tasks import SetupStaticRunTask
 from pymatgen import Composition, Lattice
 from pymatgen.core.structure import Structure
-from pymatgen.io.vaspio_set import MPGGAVaspInputSet
+from pymatgen.io.vasp.sets import MPRelaxSet
 
 __author__ = 'Anubhav Jain'
 __copyright__ = 'Copyright 2013, The Materials Project'
@@ -34,11 +34,12 @@ def structure_to_wf(structure):
     connections = defaultdict(list)  # dependencies between FireWorks
 
     # generate VASP input objects for 1st VASP run - this is put in the FW spec
-    mpvis = MPGGAVaspInputSet(user_incar_settings={'NPAR': 2})
-    incar = mpvis.get_incar(structure)
-    poscar = mpvis.get_poscar(structure)
-    kpoints = mpvis.get_kpoints(structure)
-    potcar = mpvis.get_potcar(structure)
+    mpvis = MPRelaxSet(structure, user_incar_settings={'NPAR': 2, 
+                                                       "GGAU":False})
+    incar = mpvis.incar
+    poscar = mpvis.poscar
+    kpoints = mpvis.poscar
+    potcar = mpvis.poscar
 
     # serialize the VASP input objects to the FW spec
     spec = {}
@@ -51,7 +52,7 @@ def structure_to_wf(structure):
     spec['task_type'] = 'GGA optimize structure (2x) example'
 
     # set up the custodian that we want to run
-    jobs = VaspJob.double_relaxation_run('', gzipped=False)
+    jobs = VaspJob.double_relaxation_run('')
     for j in jobs: # turn off auto npar, it doesn't work for >1 node
             j.auto_npar = False
     handlers = [VaspErrorHandler(), FrozenJobErrorHandler(), MeshSymmetryErrorHandler(),

mpworks/firetasks/README.md

@@ -0,0 +1,5 @@
+# Firetasks package
+
+This package is used by the production workflow.
+
+It contains implementations of FireTasks that are connected into workflows.

mpworks/firetasks/boltztrap_tasks.py

@@ -104,7 +104,7 @@ def get_extreme(self, d, target, maximize=True, iso_cutoff=0.05, max_didx=None):
                                     max_val = val
                                     max_temp = float(t)
                                     max_dope = d['doping'][te_type][didx]
-                                    max_mu = d['mu_doping'][te_type][t][didx]
+                                    max_mu = d['mu_doping'][te_type][str(t)][didx]
 
                                 isotropic = evs['isotropic']
                 data[te_type] = {'value': max_val, 'temperature': max_temp, 'doping': max_dope, 'mu': max_mu, 'isotropic': isotropic}
@@ -121,7 +121,6 @@ def get_extreme(self, d, target, maximize=True, iso_cutoff=0.05, max_didx=None):
 
     def run_task(self, fw_spec):
         # import here to prevent import errors in bigger MPCollab
-        from mpcollab.thermoelectrics.boltztrap_TE import BoltztrapAnalyzerTE, BoltzSPB
         # get the band structure and nelect from files
         """
         prev_dir = get_loc(fw_spec['prev_vasp_dir'])
@@ -132,7 +131,7 @@ def run_task(self, fw_spec):
         bs = vr.get_band_structure(kpoints_filename=kpoints_loc)
         """
         filename = get_slug(
-            'JOB--' + fw_spec['mpsnl']['reduced_cell_formula_abc'] + '--'
+            'JOB--' + fw_spec['mpsnl'].structure.composition.reduced_formula + '--'
             + fw_spec['task_type'])
         with open(filename, 'w+') as f:
             f.write('')
@@ -177,6 +176,10 @@ def run_task(self, fw_spec):
 
             # put the data in the database
             bta = BoltztrapAnalyzer.from_files(dir)
+
+            # 8/21/15 - Anubhav removed fs_id (also see line further below, ted['boltztrap_full_fs_id'] ...)
+            # 8/21/15 - this is to save space in MongoDB, as well as non-use of full Boltztrap output (vs rerun)
+            """
             data = bta.as_dict()
             data.update(get_meta_from_structure(bs._structure))
             data['snlgroup_id'] = fw_spec['snlgroup_id']
@@ -188,26 +191,25 @@ def run_task(self, fw_spec):
             del data['hall']  # remove because it is too large and not useful
             fs = gridfs.GridFS(tdb, "boltztrap_full_fs")
             btid = fs.put(json.dumps(jsanitize(data)))
+            """
 
             # now for the "sanitized" data
-            te_analyzer = BoltztrapAnalyzerTE.from_BoltztrapAnalyzer(bta)
-
-            ted = te_analyzer.as_dict()
+            ted = bta.as_dict()
             del ted['seebeck']
             del ted['hall']
             del ted['kappa']
             del ted['cond']
 
-            ted['boltztrap_full_fs_id'] = btid
+            # ted['boltztrap_full_fs_id'] = btid
             ted['snlgroup_id'] = fw_spec['snlgroup_id']
             ted['run_tags'] = fw_spec['run_tags']
-            ted['snl'] = fw_spec['mpsnl']
+            ted['snl'] = fw_spec['mpsnl'].as_dict()
             ted['dir_name_full'] = dir
             ted['dir_name'] = get_block_part(dir)
             ted['task_id'] = m_task['task_id']
 
-            ted['pf_doping'] = te_analyzer.get_power_factor(tau=self.TAU).as_dict()
-            ted['zt_doping'] = te_analyzer.get_ZT(kappal=self.KAPPAL, tau=self.TAU).as_dict()
+            ted['pf_doping'] = bta.get_power_factor(output='tensor', relaxation_time=self.TAU)
+            ted['zt_doping'] = bta.get_zt(output='tensor', relaxation_time=self.TAU, kl=self.KAPPAL)
 
             ted['pf_eigs'] = self.get_eigs(ted, 'pf_doping')
             ted['pf_best'] = self.get_extreme(ted, 'pf_eigs')
@@ -231,7 +233,8 @@ def run_task(self, fw_spec):
             ted['kappa_best_dope19'] = self.get_extreme(ted, 'kappa_eigs', maximize=False, max_didx=4)
 
             try:
-                bzspb = BoltzSPB(te_analyzer)
+	        from mpcollab.thermoelectrics.boltztrap_TE import BoltzSPB
+                bzspb = BoltzSPB(ted)
                 maxpf_p = bzspb.get_maximum_power_factor('p', temperature=0, tau=1E-14, ZT=False, kappal=0.5,\
                     otherprops=('get_seebeck_mu_eig', 'get_conductivity_mu_eig', \
                                                     'get_thermal_conductivity_mu_eig', 'get_average_eff_mass_tensor_mu'))
@@ -278,8 +281,8 @@ def run_task(self, fw_spec):
             update_spec = {'prev_vasp_dir': fw_spec['prev_vasp_dir'],
                        'boltztrap_dir': os.getcwd(),
                        'prev_task_type': fw_spec['task_type'],
-                       'mpsnl': fw_spec['mpsnl'],
+                       'mpsnl': fw_spec['mpsnl'].as_dict(),
                        'snlgroup_id': fw_spec['snlgroup_id'],
                        'run_tags': fw_spec['run_tags'], 'parameters': fw_spec.get('parameters')}
 
-        return FWAction(update_spec=update_spec)
+        return FWAction(update_spec=update_spec)

mpworks/firetasks/controller_tasks.py

@@ -55,7 +55,7 @@ def run_task(self, fw_spec):
 
         type_name = 'GGA+U' if 'GGA+U' in fw_spec['prev_task_type'] else 'GGA'
 
-        snl = StructureNL.from_dict(fw_spec['mpsnl'])
+        snl = fw_spec['mpsnl']
         f = Composition(snl.structure.composition.reduced_formula).alphabetical_formula
 
         fws = []
@@ -154,7 +154,7 @@ def run_task(self, fw_spec):
             print 'Adding more runs...'
             type_name = 'GGA+U' if 'GGA+U' in fw_spec['prev_task_type'] else 'GGA'
 
-            snl = StructureNL.from_dict(fw_spec['mpsnl'])
+            snl = fw_spec['mpsnl']
             f = Composition(snl.structure.composition.reduced_formula).alphabetical_formula
 
             fws = []
@@ -223,4 +223,4 @@ class DummyLegacyTask(FireTaskBase, FWSerializable):
     _fw_name = "Dummy Legacy Task"
 
     def run_task(self, fw_spec):
-        pass
+        pass