def test_init_tuple(self):
t = (1, 2, 3, 4)
pl = DataContainer(t)
self.assertEqual(len(pl), len(t),
"not the same length as source tuple")
self.assertEqual(tuple(pl.values()), t,
"conversion to tuple not the same as source tuple")
def test_update_blacklist(self):
"""Wrapping nested mapping should only apply to types not in the blacklist."""
pl = DataContainer()
pl.update([{
"a": 1,
"b": 2
}, [{
"c": 3,
"d": 4
}]],
wrap=True,
blacklist=(dict, ))
self.assertTrue(isinstance(pl[0], dict),
"nested dict wrapped, even if black listed")
self.assertTrue(isinstance(pl[1][0], dict),
"nested dict wrapped, even if black listed")
pl.clear()
pl.update({
"a": [1, 2, 3],
"b": {
"c": [4, 5, 6]
}
},
wrap=True,
blacklist=(list, ))
self.assertTrue(isinstance(pl.a, list),
"nested list wrapped, even if black listed")
self.assertTrue(isinstance(pl.b.c, list),
"nested list wrapped, even if black listed")
pl.clear()
def test_wrap_hdf(self):
"""DataContainer should be able to be initialized by HDF objects."""
h = self.project.create_hdf(self.project.path, "wrap_test")
h["foo"] = 42
h.create_group("bar")["test"] = 23
h["bar"].create_group("nested")["test"] = 23
d = DataContainer(h)
self.assertTrue(isinstance(d.bar, DataContainer),
"HDF group not wrapped from ProjectHDFio.")
self.assertTrue(isinstance(d.bar.nested, DataContainer),
"Nested HDF group not wrapped from ProjectHDFio.")
self.assertEqual(
d.foo, 42,
"Top-level node not correctly wrapped from ProjectHDFio.")
self.assertEqual(
d.bar.test, 23,
"Nested node not correctly wrapped from ProjectHDFio.")
self.assertEqual(
d.bar.nested.test, 23,
"Nested node not correctly wrapped from ProjectHDFio.")
h = h5py.File(h.file_name)
d = DataContainer(h)
self.assertTrue(isinstance(d.wrap_test.bar, DataContainer),
"HDF group not wrapped from h5py.File.")
self.assertTrue(isinstance(d.wrap_test.bar.nested, DataContainer),
"Nested HDF group not wrapped from h5py.File.")
self.assertEqual(
d.wrap_test.foo, h["wrap_test/foo"],
"Top-level node not correctly wrapped from h5py.File.")
self.assertEqual(d.wrap_test.bar.test, h["wrap_test/bar/test"],
"Nested node not correctly wrapped from h5py.File.")
self.assertEqual(d.wrap_test.bar.nested.test,
h["wrap_test/bar/nested/test"],
"Nested node not correctly wrapped from h5py.File.")
def __init__(self, project, job_name):
super(ScriptJob, self).__init__(project, job_name)
self.__version__ = "0.1"
self.__hdf_version__ = "0.2.0"
self.__name__ = "Script"
self._script_path = None
self.input = DataContainer(table_name="custom_dict")
def test_init_dict(self):
d = {"foo": 23, "test case": "bar"}
pl = DataContainer(d)
self.assertEqual(tuple(pl.items()), tuple(d.items()),
"source dict items not preserved")
with self.assertRaises(ValueError,
msg="no ValueError on invalid initializer"):
DataContainer({2: 0, 1: 1})
def store_custom_output_dict(output_dict):
folder = Path(".").cwd().parts[-1]
hdf_file = Path(".").cwd().parents[1] / folder
hdf_file = str(hdf_file) + ".h5"
hdf = FileHDFio(hdf_file)
hdf[folder].create_group("output")
obj = DataContainer(output_dict)
obj.to_hdf(hdf[folder + "/output"])
def test_read_write_consistency(self):
"""Writing a datacontainer, then reading it back in, should leave it unchanged."""
fn = "pl.yml"
self.pl.write(fn)
pl = DataContainer()
pl.read(fn)
self.assertEqual(
self.pl, pl,
"Read container from yaml, is not the same as written.")
os.remove(fn)
def test_subclass_preservation(self):
self.pl.subclass = Sub(table_name='subclass')
self.pl.to_hdf(hdf=self.hdf)
loaded = DataContainer(table_name="input")
loaded.from_hdf(hdf=self.hdf)
self.assertIsInstance(
loaded.subclass, Sub, f"Subclass not preserved on loading. "
f"Expected {Sub.__name__} but got {type(loaded.subclass).__name__}."
)
self.pl.pop('subclass')
def setUpClass(cls):
super().setUpClass()
cls.pl = DataContainer([{
"foo": "bar"
}, 2, 42, {
"next": [0, {
"depth": 23
}]
}],
table_name="input")
cls.pl["tail"] = DataContainer([2, 4, 8])
cls.hdf = cls.project.create_hdf(cls.project.path, "test")
def test_hdf_complex_members(self):
"""Values that implement to_hdf/from_hdf, should write themselves to the HDF file correctly."""
pl = DataContainer(table_name="complex")
pl.append(
self.project.create_job(self.project.job_type.ScriptJob, "dummy1"))
pl.append(
self.project.create_job(self.project.job_type.ScriptJob, "dummy2"))
pl.append(42)
pl["foo"] = "bar"
pl.to_hdf(hdf=self.hdf)
pl2 = self.hdf["complex"].to_object()
self.assertEqual(type(pl[0]), type(pl2[0]))
self.assertEqual(type(pl[1]), type(pl2[1]))
def setUpClass(cls):
super().setUpClass()
cls.hdf = cls.project.create_hdf(cls.project.path, "hdf")
cls.hdf["number"] = 42
cls.hdf["array"] = np.arange(100)
cls.data = DataContainer([1, 2, "three", 4.0])
cls.data.to_hdf(cls.hdf, "data")
def test_to_hdf_type(self):
"""Should write correct type information."""
self.pl.to_hdf(hdf=self.hdf)
self.assertEqual(self.hdf["input/NAME"], "DataContainer")
self.assertEqual(self.hdf["input/OBJECT"], "DataContainer")
self.assertEqual(
self.hdf["input/TYPE"],
"<class 'pyiron_base.generic.datacontainer.DataContainer'>")
h = self.hdf.open('nested')
pl = DataContainer(self.pl)
pl.to_hdf(hdf=h)
self.assertEqual(h["NAME"], "DataContainer")
self.assertEqual(h["OBJECT"], "DataContainer")
self.assertEqual(
h["TYPE"],
"<class 'pyiron_base.generic.datacontainer.DataContainer'>")
def test_create_group(self):
"""create_group should not erase existing groups."""
cont = DataContainer()
sub1 = cont.create_group("sub")
self.assertTrue(isinstance(sub1, DataContainer),
"create_group doesn't return DataContainer")
sub1.foo = 42
sub2 = cont.create_group("sub")
self.assertEqual(sub1.foo, sub2.foo,
"create_group overwrites existing data.")
self.assertTrue(
sub1 is sub2,
"create_group return new DataContainer group instead of existing one."
)
with self.assertRaises(
ValueError, msg="No ValueError on existing data in Container"):
sub1.create_group("foo")
def test_mark(self):
pl = DataContainer([1, 2, 3])
pl.mark(1, "foo")
self.assertEqual(pl[1], pl.foo,
"marked element does not refer to correct element")
pl.mark(2, "foo")
self.assertEqual(pl[2], pl.foo, "marking with existing key broken")
with self.assertRaises(IndexError,
msg="no IndexError on invalid index"):
pl.mark(10, "foo")
def test_hdf_no_wrap(self):
"""Nested mappings should not be wrapped as DataContainer after reading."""
l = DataContainer(table_name="mappings")
l.append({"foo": "bar"})
l.append([1, 2, 3])
l.to_hdf(self.hdf)
m = l.copy()
m.from_hdf(self.hdf, group_name="mappings")
self.assertEqual(l, m,
"List with nested mappings not restored from HDF.")
self.assertTrue(isinstance(m[0], dict),
"dicts wrapped after reading from HDF.")
self.assertTrue(isinstance(m[1], list),
"lists wrapped after reading from HDF.")
def get_custom_dict():
folder = Path(".").cwd().parts[-1]
project_folder = Path(".").cwd().parents[1]
hdf_file = project_folder / folder
hdf_file = str(hdf_file).replace("\\", "/") + ".h5"
if Path(hdf_file).exists():
obj = DataContainer()
obj.from_hdf(
hdf=FileHDFio(hdf_file), group_name=folder + "/input/custom_dict"
)
obj["project_dir"] = str(project_folder)
return obj
elif Path("input.json").exists():
with open("input.json") as f:
return json.load(f)
else:
warnings.warn("{} not found".format(hdf_file))
return None
def test_get_nested(self):
n = [{"foo": "bar"}, 2, 42, {"next": [0, {"depth": 23}]}]
pl = DataContainer(n)
self.assertEqual(type(pl[0]), DataContainer,
"nested dict not converted to DataContainer")
self.assertEqual(type(pl["3/next"]), DataContainer,
"nested list not converted to DataContainer")
self.assertEqual(type(pl["0/foo"]), str,
"nested str converted to DataContainer")
def test_insert(self):
pl = DataContainer([1, 2, 3])
pl.insert(1, 42, key="foo")
self.assertTrue(pl[0] == 1 and pl[1] == 42 and pl[2] == 2,
"insert does not properly set value")
pl.insert(1, 24, key="bar")
self.assertTrue(pl[0] == 1 and pl.bar == 24 and pl.foo == 42,
"insert does not properly update keys")
pl.insert(10, 4)
self.assertEqual(pl[-1], 4,
"insert does not handle out of bounds gracefully")
def test_del_item(self):
pl = DataContainer({0: 1, "a": 2, "foo": 3})
with self.assertRaises(ValueError,
msg="no ValueError on invalid index type"):
del pl[{}]
del pl["a"]
self.assertTrue("a" not in pl, "delitem does not delete with str key")
del pl[0]
self.assertTrue(pl[0] != 1, "delitem does not delete with index")
def test_hdf_pandas(self):
"""Values that implement to_hdf/from_hdf, should write themselves to the HDF file correctly."""
pl = DataContainer(table_name="pandas")
pl.append(pd.DataFrame({"a": [1, 2], "b": ["x", "y"]}))
pl.to_hdf(hdf=self.hdf)
pl2 = self.hdf["pandas"].to_object()
self.assertEqual(type(pl[0]), type(pl2[0]))
def test_hdf_empty_group(self):
"""Writing a list without table_name or group_name should only work if the HDF group is empty."""
l = DataContainer([1, 2, 3])
with self.assertRaises(
ValueError,
msg="No exception when writing to full hdf group."):
l.to_hdf(self.hdf)
h = self.hdf.create_group("empty_group")
l.to_hdf(h)
self.assertEqual(l, h.to_object())
def test_from_hdf_readonly(self):
"""Reading from HDF should restore the read-only property."""
self.pl.to_hdf(hdf=self.hdf, group_name="read_only_from")
pl = DataContainer()
pl.from_hdf(self.hdf, group_name="read_only_from")
self.assertEqual(pl.read_only, self.hdf["read_only_from/READ_ONLY"],
"read-only parameter not correctly read from HDF")
self.hdf["read_only_from/READ_ONLY"] = True
pl.from_hdf(self.hdf, group_name="read_only_from")
self.assertEqual(pl.read_only, self.hdf["read_only_from/READ_ONLY"],
"read-only parameter not correctly read from HDF")
def test_set_append(self):
pl = DataContainer()
# should not raise and exception
pl[0] = 1
pl[1] = 2
self.assertEqual(pl[0], 1, "append via index broken on empty list")
self.assertEqual(pl[1], 2, "append via index broken on non-empty list")
pl.append([])
self.assertTrue(
isinstance(pl[-1], list),
"append wraps sequences as DataContainer, but should not")
pl.append({})
self.assertTrue(
isinstance(pl[-1], dict),
"append wraps mappings as DataContainer, but should not")
def test_from_hdf_readonly(self):
"""Reading from HDF should restore the read-only property."""
self.pl.to_hdf(hdf=self.hdf, group_name="read_only_from")
pl = DataContainer()
pl.from_hdf(self.hdf, group_name="read_only_from")
self.assertEqual(pl.read_only, self.hdf["read_only_from/READ_ONLY"],
"read-only parameter not correctly read from HDF")
self.hdf["read_only_from/READ_ONLY"] = True
with warnings.catch_warnings(record=True) as w:
pl.from_hdf(self.hdf, group_name="read_only_from")
self.assertEqual(
len(w), 0,
"from_hdf on read_only DataContainer should not call _read_only_error."
)
self.assertEqual(pl.read_only, self.hdf["read_only_from/READ_ONLY"],
"read-only parameter not correctly read from HDF")
class ScriptJob(GenericJob):
"""
The ScriptJob class allows to submit Python scripts and Jupyter notebooks to the pyiron job management system.
Args:
project (ProjectHDFio): ProjectHDFio instance which points to the HDF5 file the job is stored in
job_name (str): name of the job, which has to be unique within the project
Simple example:
Step 1. Create the notebook to be submitted, for ex. 'example.ipynb', and save it -- Can contain any code like:
```
import json
with open('script_output.json','w') as f:
json.dump({'x': [0,1]}, f) # dump some data into a JSON file
```
Step 2. Create the submitter notebook, for ex. 'submit_example_job.ipynb', which submits 'example.ipynb' to the
pyiron job management system, which can have the following code:
```
from pyiron_base import Project
pr = Project('scriptjob_example') # save the ScriptJob in the 'scriptjob_example' project
scriptjob = pr.create.job.ScriptJob('scriptjob') # create a ScriptJob named 'scriptjob'
scriptjob.script_path = 'example.ipynb' # specify the PATH to the notebook you want to submit.
```
Step 3. Check the job table to get details about 'scriptjob' by using:
```
pr.job_table()
```
Step 4. If the status of 'scriptjob' is 'finished', load the data from the JSON file into the
'submit_example_job.ipynb' notebook by using:
```
import json
with open(scriptjob.working_directory + '/script_output.json') as f:
data = json.load(f) # load the data from the JSON file
```
More sophisticated example:
The script in ScriptJob can also be more complex, e.g. running its own pyiron calculations.
Here we show how it is leveraged to run a multi-core atomistic calculation.
Step 1. 'example.ipynb' can contain pyiron_atomistics code like:
```
from pyiron_atomistics import Project
pr = Project('example')
job = pr.create.job.Lammps('job') # we name the job 'job'
job.structure = pr.create.structure.ase_bulk('Fe') # specify structure
# Optional: get an input value from 'submit_example_job.ipynb', the notebook which submits
# 'example.ipynb'
number_of_cores = pr.data.number_of_cores
job.server.cores = number_of_cores
job.run() # run the job
# save a custom output, that can be used by the notebook 'submit_example_job.ipynb'
job['user/my_custom_output'] = 16
```
Step 2. 'submit_example_job.ipynb', can then have the following code:
```
from pyiron_base import Project
pr = Project('scriptjob_example') # save the ScriptJob in the 'scriptjob_example' project
scriptjob = pr.create.job.ScriptJob('scriptjob') # create a ScriptJob named 'scriptjob'
scriptjob.script_path = 'example.ipynb' # specify the PATH to the notebook you want to submit.
# In this example case, 'example.ipynb' is in the same
# directory as 'submit_example_job.ipynb'
# Optional: to submit the notebook to a queueing system
number_of_cores = 1 # number of cores to be used
scriptjob.server.cores = number_of_cores
scriptjob.server.queue = 'cmfe' # specify the queue to which the ScriptJob job is to be submitted
scriptjob.server.run_time = 120 # specify the runtime limit for the ScriptJob job in seconds
# Optional: save an input, such that it can be accessed by 'example.ipynb'
pr.data.number_of_cores = number_of_cores
pr.data.write()
# run the ScriptJob job
scriptjob.run()
```
Step 3. Check the job table by using:
```
pr.job_table()
```
in addition to containing details on 'scriptjob', the job_table also contains the details of the child
'job/s' (if any) that were submitted within the 'example.ipynb' notebook.
Step 4. Reload and analyse the child 'job/s': If the status of a child 'job' is 'finished', it can be loaded
into the 'submit_example_job.ipynb' notebook using:
```
job = pr.load('job') # remember in Step 1., we wanted to run a job named 'job', which has now
# 'finished'
```
this loads 'job' into the 'submit_example_job.ipynb' notebook, which can be then used for analysis,
```
job.output.energy_pot[-1] # via the auto-complete
job['user/my_custom_output'] # the custom output, directly from the hdf5 file
```
Attributes:
attribute: job_name
name of the job, which has to be unique within the project
.. attribute:: status
execution status of the job, can be one of the following [initialized, appended, created, submitted, running,
aborted, collect, suspended, refresh, busy, finished]
.. attribute:: job_id
unique id to identify the job in the pyiron database
.. attribute:: parent_id
job id of the predecessor job - the job which was executed before the current one in the current job series
.. attribute:: master_id
job id of the master job - a meta job which groups a series of jobs, which are executed either in parallel or in
serial.
.. attribute:: child_ids
list of child job ids - only meta jobs have child jobs - jobs which list the meta job as their master
.. attribute:: project
Project instance the jobs is located in
.. attribute:: project_hdf5
ProjectHDFio instance which points to the HDF5 file the job is stored in
.. attribute:: job_info_str
short string to describe the job by it is job_name and job ID - mainly used for logging
.. attribute:: working_directory
working directory of the job is executed in - outside the HDF5 file
.. attribute:: path
path to the job as a combination of absolute file system path and path within the HDF5 file.
.. attribute:: version
Version of the hamiltonian, which is also the version of the executable unless a custom executable is used.
.. attribute:: executable
Executable used to run the job - usually the path to an external executable.
.. attribute:: library_activated
For job types which offer a Python library pyiron can use the python library instead of an external executable.
.. attribute:: server
Server object to handle the execution environment for the job.
.. attribute:: queue_id
the ID returned from the queuing system - it is most likely not the same as the job ID.
.. attribute:: logger
logger object to monitor the external execution and internal pyiron warnings.
.. attribute:: restart_file_list
list of files which are used to restart the calculation from these files.
.. attribute:: job_type
Job type object with all the available job types: ['ExampleJob', 'SerialMaster', 'ParallelMaster', 'ScriptJob',
'ListMaster']
.. attribute:: script_path
the absolute path to the python script
"""
def __init__(self, project, job_name):
super(ScriptJob, self).__init__(project, job_name)
self.__version__ = "0.1"
self.__hdf_version__ = "0.2.0"
self.__name__ = "Script"
self._script_path = None
self.input = DataContainer(table_name="custom_dict")
@property
def script_path(self):
"""
Python script path
Returns:
str: absolute path to the python script
"""
return self._script_path
@script_path.setter
def script_path(self, path):
"""
Python script path
Args:
path (str): relative or absolute path to the python script or a corresponding notebook
"""
if isinstance(path, str):
self._script_path = self._get_abs_path(path)
self.executable = self._executable_command(
working_directory=self.working_directory,
script_path=self._script_path)
else:
raise TypeError("path should be a string, but ", path, " is a ",
type(path), " instead.")
def validate_ready_to_run(self):
if self.script_path is None:
raise TypeError(
'ScriptJob.script_path expects a path but got None. Please provide a path before '
+ 'running.')
def set_input_to_read_only(self):
"""
This function enforces read-only mode for the input classes, but it has to be implement in the individual
classes.
"""
self.input.read_only = True
def to_hdf(self, hdf=None, group_name=None):
"""
Store the ScriptJob in an HDF5 file
Args:
hdf (ProjectHDFio): HDF5 group object - optional
group_name (str): HDF5 subgroup name - optional
"""
super(ScriptJob, self).to_hdf(hdf=hdf, group_name=group_name)
with self.project_hdf5.open("input") as hdf5_input:
hdf5_input["path"] = self._script_path
self.input.to_hdf(hdf5_input)
def from_hdf(self, hdf=None, group_name=None):
"""
Restore the ScriptJob from an HDF5 file
Args:
hdf (ProjectHDFio): HDF5 group object - optional
group_name (str): HDF5 subgroup name - optional
"""
super(ScriptJob, self).from_hdf(hdf=hdf, group_name=group_name)
if "HDF_VERSION" in self.project_hdf5.list_nodes():
version = self.project_hdf5["HDF_VERSION"]
else:
version = "0.1.0"
if version == "0.1.0":
with self.project_hdf5.open("input") as hdf5_input:
try:
self.script_path = hdf5_input["path"]
gp = GenericParameters(table_name="custom_dict")
gp.from_hdf(hdf5_input)
for k in gp.keys():
self.input[k] = gp[k]
except TypeError:
pass
elif version == "0.2.0":
with self.project_hdf5.open("input") as hdf5_input:
try:
self.script_path = hdf5_input["path"]
except TypeError:
pass
self.input.from_hdf(hdf5_input)
else:
raise ValueError("Cannot handle hdf version: {}".format(version))
def write_input(self):
"""
Copy the script to the working directory - only python scripts and jupyter notebooks are supported
"""
if self._script_path is not None:
file_name = os.path.basename(self._script_path)
shutil.copyfile(src=self._script_path,
dst=os.path.join(self.working_directory,
file_name))
def collect_output(self):
"""
Collect output function updates the master ID entries for all the child jobs created by this script job, if the
child job is already assigned to an master job nothing happens - master IDs are not overwritten.
"""
for job in self.project.iter_jobs(recursive=False,
convert_to_object=False):
pr_job = self.project.open(
os.path.relpath(job.working_directory, self.project.path))
for subjob_id in pr_job.get_job_ids(recursive=False):
if pr_job.db.get_item_by_id(subjob_id)["masterid"] is None:
pr_job.db.item_update({"masterid": str(job.job_id)},
subjob_id)
def run_if_lib(self):
"""
Compatibility function - but library run mode is not available
"""
raise NotImplementedError(
"Library run mode is not implemented for script jobs.")
def collect_logfiles(self):
"""
Compatibility function - but no log files are being collected
"""
pass
@staticmethod
def _executable_command(working_directory, script_path):
"""
internal function to generate the executable command to either use jupyter or python
Args:
working_directory (str): working directory of the current job
script_path (str): path to the script which should be executed in the working directory
Returns:
str: executable command
"""
file_name = os.path.basename(script_path)
path = os.path.join(working_directory, file_name)
if file_name[-6:] == ".ipynb":
return (
"jupyter nbconvert --ExecutePreprocessor.timeout=9999999 --to notebook --execute "
+ path)
elif file_name[-3:] == ".py":
return "python " + path
else:
raise ValueError("Filename not recognized: ", path)
def _executable_activate_mpi(self):
"""
Internal helper function to switch the executable to MPI mode
"""
pass
@staticmethod
def _get_abs_path(path):
"""
internal function to convert absolute or relative paths to absolute paths, using os.path.normpath,
os.path.abspath and os.path.curdir
Args:
path (str): relative or absolute path
Returns:
str: absolute path
"""
return os.path.normpath(
os.path.join(os.path.abspath(os.path.curdir), path))
def test_init_set(self):
s = {1, 2, 3, 4}
pl = DataContainer(s)
self.assertEqual(len(pl), len(s), "not the same length as source set")
self.assertEqual(set(pl.values()), s,
"conversion to set not the same as source set")
def test_recursive_append(self):
input_tmp = DataContainer()
input_tmp['some/argument/inside/another/argument'] = 3
self.assertEqual(input_tmp['some/argument/inside/another/argument'], 3)
self.assertEqual(input_tmp.some.argument.inside.another.argument, 3)
self.assertEqual(type(input_tmp.some), DataContainer)
def test_from_hdf_group(self):
"""Reading from HDF should give back the same list as written even with custom group name."""
self.pl.to_hdf(hdf=self.hdf, group_name="test_group")
l = DataContainer(table_name="input")
l.from_hdf(hdf=self.hdf, group_name="test_group")
self.assertEqual(self.pl, l)
def test_hdf_empty_list(self):
"""Writing and reading an empty list should work."""
l = DataContainer(table_name="empty_list")
l.to_hdf(self.hdf)
l.from_hdf(self.hdf)
self.assertEqual(len(l), 0, "Empty list read from HDF not empty.")
def test_init_list(self):
l = [1, 2, 3, 4]
pl = DataContainer(l)
self.assertEqual(len(pl), len(l), "not the same length as source list")
self.assertEqual(list(pl.values()), l,
"conversion to list not the same as source list")
