Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

topology featurizers debug #460

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

topology featurizers debug #460

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
eabe0dc
topology featurizers debug
XiaoqZhang Aug 14, 2024
376c598
ci: update .deepsource.toml
deepsource-io[bot] Aug 14, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
24 changes: 12 additions & 12 deletions src/mofdscribe/featurizers/topology/_tda_helpers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,7 @@

from mofdscribe.featurizers.utils import flat
from mofdscribe.featurizers.utils.aggregators import MA_ARRAY_AGGREGATORS
from mofdscribe.featurizers.utils.substructures import filter_element
from mofdscribe.featurizers.utils.substructures import filter_element_for_ph


# @np_cache
Expand Down Expand Up @@ -104,8 +104,8 @@ def make_supercell(
xyz_periodic_copies = []
element_copies = []

# xyz_periodic_copies.append(coords)
# element_copies.append(np.array(elements).reshape(-1,1))
xyz_periodic_copies.append(coords)
element_copies.append(np.array(elements).reshape(-1,1))
Comment on lines +107 to +108
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

ok, seems like a good catch, because the x==y==z==0 condition is skipped atm

min_range = -3 # we aren't going in the minimum direction too much, so can make this small
max_range = 20 # make this large enough, but can modify if wanting an even larger cell

Expand Down Expand Up @@ -228,7 +228,7 @@ def get_images(
# ToDo: only do this for all if we want
def get_persistent_images_for_structure(
structure: Structure,
elements: List[List[str]],
elements: List[str],
compute_for_all_elements: bool = True,
min_size: int = 20,
spread: float = 0.2,
Expand All @@ -245,7 +245,7 @@ def get_persistent_images_for_structure(

Args:
structure (Structure): input structure
elements (List[List[str]]): list of elements to compute for
elements (List[str]): list of element groups to compute for
compute_for_all_elements (bool): compute for all elements
min_size (int): minimum size of the cell for construction of persistent images
spread (float): spread of kernel for construction
Expand Down Expand Up @@ -273,9 +273,9 @@ def get_persistent_images_for_structure(
specs = []
for mb, mp in zip(max_b, max_p):
specs.append({"minBD": 0, "maxB": mb, "maxP": mp})
for element in elements:
for elements_group in elements:
try:
filtered_structure = filter_element(structure, element)
filtered_structure = filter_element_for_ph(structure, elements_group)
coords, _weights, _elements = _coords_for_structure(
filtered_structure,
min_size=min_size,
Expand All @@ -294,7 +294,7 @@ def get_persistent_images_for_structure(
dimensions=(0, 1, 2),
)
except Exception:
logger.exception(f"Error computing persistent images for {element}")
logger.exception(f"Error computing persistent images for {elements_group}")
images = {}
for dim in [0, 1, 2]:
im = np.zeros((pixels[0], pixels[1]))
Expand All @@ -304,8 +304,8 @@ def get_persistent_images_for_structure(
persistent_dia[:] = np.nan

# ToDo: make sure that we have the correct length
element_images["image"][element] = images
element_images["array"][element] = persistent_dia
element_images["image"][elements_group] = images
element_images["array"][elements_group] = persistent_dia

if compute_for_all_elements:
try:
Expand Down Expand Up @@ -391,7 +391,7 @@ def get_diagrams_for_structure(
nan_array[:] = np.nan
for element in elements:
try:
filtered_structure = filter_element(structure, element)
filtered_structure = filter_element_for_ph(structure, element)
coords, weights, _elements = _coords_for_structure(
filtered_structure,
min_size=min_size,
Expand Down Expand Up @@ -442,7 +442,7 @@ def get_persistence_image_limits_for_structure(
limits = defaultdict(list)
for element in elements:
try:
filtered_structure = filter_element(structure, element)
filtered_structure = filter_element_for_ph(structure, element)

coords, weights, _elements = _coords_for_structure(
filtered_structure,
Expand Down
17 changes: 9 additions & 8 deletions src/mofdscribe/featurizers/topology/ph_hist.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,14 +34,14 @@ class PHHist(MOFBaseFeaturizer):

def __init__(
self,
atom_types: Tuple[str] = (
atom_types: List[str] = [
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tuple was originally used because it is immutable and hence a better default for arguments.

But since we do not change the atom_types this is probably a safe choice

"C-H-N-O",
"F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V"
"-Ag-Nd-U-Ba-Ce-K-Ga-Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-"
"Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-"
"Th-Np-Lu-Rh-Pu",
),
],
compute_for_all_elements: bool = True,
dimensions: Tuple[int] = (1, 2),
min_size: int = 20,
Expand All @@ -57,12 +57,12 @@ def __init__(
"""Initialize the PHStats object.

Args:
atom_types (tuple): Atoms that are used to create substructures
atom_types (list): Atoms that are used to create substructures
for which the persistent homology statistics are computed.
Defaults to ( "C-H-N-O", "F-Cl-Br-I",
Defaults to [ "C-H-N-O", "F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V-Ag-Nd-U-Ba-Ce-K-Ga-
Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ).
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ].
compute_for_all_elements (bool): Compute descriptor for original structure with all atoms.
Defaults to True.
dimensions (Tuple[int]): Dimensions of topological features to consider.
Expand All @@ -89,9 +89,10 @@ def __init__(
"""
atom_types = [] if atom_types is None else atom_types
self.elements = atom_types
self.atom_types = (
list(atom_types) + ["all"] if compute_for_all_elements else list(atom_types)
)
if compute_for_all_elements:
self.atom_types = atom_types + ["all"]
else:
self.atom_types = atom_types
self.compute_for_all_elements = compute_for_all_elements
self.dimensions = dimensions
self.min_size = min_size
Expand Down
20 changes: 10 additions & 10 deletions src/mofdscribe/featurizers/topology/ph_image.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -53,14 +53,14 @@ class PHImage(MOFBaseFeaturizer):

def __init__(
self,
atom_types: Optional[Tuple[str]] = (
atom_types: Optional[List[str]] = [
"C-H-N-O",
"F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V"
"-Ag-Nd-U-Ba-Ce-K-Ga-Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-"
"Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-"
"Th-Np-Lu-Rh-Pu",
),
"Th-Np-Lu-Rh-Pu"
],
dimensions: Tuple[int] = (0, 1, 2),
compute_for_all_elements: bool = True,
min_size: int = 20,
Expand All @@ -78,14 +78,14 @@ def __init__(
"""Construct a PHImage object.

Args:
atom_types (Tuple[str], optional): Atoms that are used to create
atom_types (List[str], optional): Atoms that are used to create
substructures that are analysed using persistent homology.
If multiple atom types separated by hash are provided, e.g.
"C-H-N-O", then the substructure consists of all atoms of type
C, H, N, or O. Defaults to ( "C-H-N-O", "F-Cl-Br-I",
C, H, N, or O. Defaults to [ "C-H-N-O", "F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V-Ag-Nd-U-Ba-
Ce-K-Ga-Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-Ho-Re-Be-Rb-La-Sn-Cs-Pb-
Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu").
Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu"].
dimensions (Tuple[int]): Dimensions of topological
features to consider for persistence images. Defaults to (0, 1, 2).
compute_for_all_elements (bool): If true, compute
Expand Down Expand Up @@ -181,14 +181,14 @@ def get_birth_persistance_death_from_pixel(self, dimension: int, x: int, y: int)

def _get_feature_labels(self) -> List[str]:
labels = []
_elements = list(self.atom_types)
_elements = self.atom_types
if self.compute_for_all_elements:
_elements.append("all")
for element in _elements:
for elements_group in _elements:
for dim in self.dimensions:
for pixel_a in range(self.image_size[0]):
for pixel_b in range(self.image_size[1]):
labels.append(f"phimage_{element}_{dim}_{pixel_a}_{pixel_b}")
labels.append(f"phimage_{elements_group}_{dim}_{pixel_a}_{pixel_b}")

return labels

Expand Down Expand Up @@ -293,7 +293,7 @@ def _featurize(
alpha_weighting=self.alpha_weight,
)
features = []
elements = list(self.atom_types)
elements = self.atom_types
if self.compute_for_all_elements:
elements.append("all")
for element in elements:
Expand Down
17 changes: 9 additions & 8 deletions src/mofdscribe/featurizers/topology/ph_stats.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,14 +34,14 @@ class PHStats(MOFBaseFeaturizer):

def __init__(
self,
atom_types: Tuple[str] = (
atom_types: List[str] = [
"C-H-N-O",
"F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V"
"-Ag-Nd-U-Ba-Ce-K-Ga-Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-"
"Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-"
"Th-Np-Lu-Rh-Pu",
),
],
compute_for_all_elements: bool = True,
dimensions: Tuple[int] = (1, 2),
min_size: int = 20,
Expand All @@ -54,12 +54,12 @@ def __init__(
"""Initialize the PHStats object.

Args:
atom_types (tuple): Atoms that are used to create substructures
atom_types (list): Atoms that are used to create substructures
for which the persistent homology statistics are computed.
Defaults to ( "C-H-N-O", "F-Cl-Br-I",
Defaults to [ "C-H-N-O", "F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V-Ag-Nd-U-Ba-Ce-K-Ga-
Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ).
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ].
compute_for_all_elements (bool): Compute descriptor for original structure with all atoms.
Defaults to True.
dimensions (Tuple[int]): Dimensions of topological features to consider.
Expand All @@ -81,9 +81,10 @@ def __init__(
"""
atom_types = [] if atom_types is None else atom_types
self.elements = atom_types
self.atom_types = (
list(atom_types) + ["all"] if compute_for_all_elements else list(atom_types)
)
if compute_for_all_elements:
self.atom_types = atom_types + ["all"]
else:
self.atom_types = atom_types
self.compute_for_all_elements = compute_for_all_elements
self.dimensions = dimensions
self.min_size = min_size
Expand Down
17 changes: 9 additions & 8 deletions src/mofdscribe/featurizers/topology/ph_vect.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -140,14 +140,14 @@ class PHVect(MOFBaseFeaturizer):

def __init__(
self,
atom_types: Tuple[str] = (
atom_types: List[str] = [
"C-H-N-O",
"F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V"
"-Ag-Nd-U-Ba-Ce-K-Ga-Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-"
"Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-Hf-Ir-Nb-Pd-Hg-"
"Th-Np-Lu-Rh-Pu",
),
],
compute_for_all_elements: bool = True,
dimensions: Tuple[int] = (1, 2),
min_size: int = 20,
Expand All @@ -165,14 +165,14 @@ def __init__(
"""Construct a PHVect instance.

Args:
atom_types (tuple): Atoms that are used to create substructures
atom_types (list): Atoms that are used to create substructures
that are analysed using persistent homology.
If multiple atom types separated by hash are provided,
e.g. "C-H-N-O", then the substructure consists of all atoms of type C, H, N, or O.
Defaults to ( "C-H-N-O", "F-Cl-Br-I",
Defaults to [ "C-H-N-O", "F-Cl-Br-I",
"Cu-Mn-Ni-Mo-Fe-Pt-Zn-Ca-Er-Au-Cd-Co-Gd-Na-Sm-Eu-Tb-V-Ag-Nd-U-Ba-Ce-K-Ga-
Cr-Al-Li-Sc-Ru-In-Mg-Zr-Dy-W-Yb-Y-Ho-Re-Be-Rb-La-Sn-Cs-Pb-Pr-Bi-Tm-Sr-Ti-
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ).
Hf-Ir-Nb-Pd-Hg-Th-Np-Lu-Rh-Pu", ].
compute_for_all_elements (bool): If true, compute persistence images
for full structure (i.e. with all elements). If false, it will only do it
for the substructures specified with `atom_types`. Defaults to True.
Expand Down Expand Up @@ -209,9 +209,10 @@ def __init__(
"""
atom_types = [] if atom_types is None else atom_types
self.elements = atom_types
self.atom_types = (
list(atom_types) + ["all"] if compute_for_all_elements else list(atom_types)
)
if compute_for_all_elements:
self.atom_types = atom_types + ["all"]
else:
self.atom_types = atom_types
self.compute_for_all_elements = compute_for_all_elements
self.min_size = min_size
self.dimensions = dimensions
Expand Down
33 changes: 33 additions & 0 deletions src/mofdscribe/featurizers/utils/substructures.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -35,6 +35,39 @@ def filter_element(
return Structure.from_sites(keep_sites)
else: # input is molecule or IMolecule
return Molecule.from_sites(keep_sites)


def filter_element_for_ph(
Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

issue (complexity): Consider refactoring to consolidate logic into a single function.

The new code introduces unnecessary complexity by duplicating logic in the filter_element_for_ph function, which increases maintenance overhead. The use of elements_group as a tuple doesn't add functional value and makes the code harder to read. Additionally, the handling of elements is inconsistent, which could lead to bugs. Consider refactoring to consolidate the logic into a single function, using list comprehensions for clarity, and ensuring consistent handling of elements. This will improve readability and maintainability.

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

can you make a suggestion?

Copy link

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Certainly! To address the complexity issue, you could refactor the filter_element and filter_element_for_ph functions to share common logic. Here's a suggestion:

  1. Create a Helper Function: Extract the common logic into a helper function that both filter_element and filter_element_for_ph can call. This will reduce duplication and make the code easier to maintain.

  2. Use List Comprehensions: If possible, use list comprehensions to make the filtering logic more concise and readable.

  3. Consistent Handling of elements: Ensure that the handling of elements is consistent across both functions. This might involve standardizing the input format or processing steps.

Here's a rough example of how you might structure the refactoring:

def filter_elements_common(sites, elements, elements_group):
    return [site for site in sites if site.specie.symbol in elements or site.specie.symbol in elements_group]

def filter_element(structure, elements, elements_group=()):
    keep_sites = filter_elements_common(structure.sites, elements, elements_group)
    return Structure.from_sites(keep_sites) if isinstance(structure, Structure) else Molecule.from_sites(keep_sites)

def filter_element_for_ph(structure, elements, elements_group=()):
    # If there's any specific logic for persistence homology, handle it here
    keep_sites = filter_elements_common(structure.sites, elements, elements_group)
    return Structure.from_sites(keep_sites) if isinstance(structure, Structure) else Molecule.from_sites(keep_sites)

This approach keeps the core filtering logic in one place, making it easier to update and maintain. If filter_element_for_ph requires additional specific logic, you can still handle that separately within its function. Let me know if you need further clarification or assistance!

structure: Union[Structure, IStructure, Molecule, IMolecule], elements: str
) -> Structure:
"""Filter a structure by element.

Args:
structure (Union[Structure, IStructure, Molecule, IMolecule]): input structure
elements (str): element to filter

Returns:
filtered_structure (Structure): filtered structure
"""
elements_ = []
elements_group = (elements,)
for atom_type in elements_group:
Comment on lines +52 to +54
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

i'm not sure I understand this change. You are certainly right that the old code behaves in an unexpected way for elements with more than two symbols in the letter. But wouldn't then the clearer fix be something else? For example, always split by - first and then handle other things

if "-" in atom_type:
elements_.extend(atom_type.split("-"))
else:
elements_.append(atom_type)
keep_sites = []
for site in structure.sites:
if site.specie.symbol in elements_:
keep_sites.append(site)
if len(keep_sites) == 0:
return None

input_is_structure = isinstance(structure, (Structure, IStructure))
if input_is_structure:
return Structure.from_sites(keep_sites)
else: # input is molecule or IMolecule
return Molecule.from_sites(keep_sites)


def elements_in_structure(structure: Structure) -> List[str]:
Expand Down