mandos.analysis.prepping

Complexity

Total Complexity

11

Size/Duplication

Total Lines	86
Duplicated Lines	0 %

Importance

Changes

0

3 Methods

Rating	Name	Duplication	Size	Complexity
A	MatrixPrep.ecfp_matrix()	0	23	4
A	MatrixPrep.from_files()	0	11	3
A	MatrixPrep.create()	0	18	4

"""
X.
"""
from __future__ import annotations
from dataclasses import dataclass
from pathlib import Path
from typing import TypeVar, Mapping, Sequence, List

import numpy as np

Unable to import 'numpy'

import pandas as pd

Unable to import 'pandas'

from typeddfs import BaseDf

Unable to import 'typeddfs'

from typeddfs.df_errors import UnsupportedOperationError

Unable to import 'typeddfs.df_errors'

from mandos import logger
from mandos.analysis.io_defns import SimilarityDfLongForm, SimilarityDfShortForm
from mandos.entries.searcher import InputFrame
from mandos.model.rdkit_utils import RdkitUtils

T = TypeVar("T", bound=BaseDf)

Class name "T" doesn't conform to PascalCase naming style ('[^\\W\\da-z][^\\W_]+$' pattern)

@dataclass(frozen=True, repr=True)

Missing class docstring

class MatrixPrep:
    kind: str
    normalize: bool
    log: bool
    invert: bool

    def from_files(self, paths: Sequence[Path]) -> SimilarityDfLongForm:

Missing function or method docstring

        dct = {}
        for p in paths:

Variable name "p" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

            key = p.with_suffix("").name
            try:
                mx = SimilarityDfShortForm.read_file(p)

Variable name "mx" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

                dct[key] = mx
            except (OSError, UnsupportedOperationError, ValueError):
                logger.error(f"Failed to load matrix at {str(p)}")
                raise
        return self.create(dct)

    def create(self, key_to_mx: Mapping[str, SimilarityDfShortForm]) -> SimilarityDfLongForm:

Missing function or method docstring

        df = SimilarityDfLongForm(

Variable name "df" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

            pd.concat([mx.to_long_form(self.kind, key) for key, mx in key_to_mx.items()])
        )
        vals = df["value"]
        if self.invert:
            vals = -vals
        if self.normalize:
            mn, mx = vals.min(), vals.max()

Variable name "mn" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

Variable name "mx" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

            vals = (vals - mn) / (mn - mx)
        if self.log:
            # this is a bit stupid, but calc the log then normalize again
            # we can't take the log before normalization because we might have negative values
            vals = vals.map(np.log10)
            mn, mx = vals.min(), vals.max()

Variable name "mn" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

Variable name "mx" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

            vals = (vals - mn) / (mn - mx)
        df["value"] = vals
        return SimilarityDfLongForm.convert(df)

    @classmethod
    def ecfp_matrix(cls, df: InputFrame, radius: int, n_bits: int) -> SimilarityDfShortForm:

This function exceeds the maximum number of variables (17/15).

Missing function or method docstring

Argument name "df" doesn't conform to snake_case naming style ('([^\\W\\dA-Z][^\\WA-Z]2,|_[^\\WA-Z]*|__[^\\WA-Z\\d_][^\\WA-Z]+__)$' pattern)

        # TODO: This is inefficient and long

TODO and FIXME comments should generally be avoided.

        indices = range(len(df))
        keys = df["inchikey"]
        on_bits = [
            RdkitUtils.ecfp(c, radius=radius, n_bits=n_bits).list_on for c in df.get_structures()
        ]
        the_rows: List[List[float]] = []
        for i, row_key, row_print in zip(indices, keys, on_bits):

The variable row_key seems to be unused.

            for j, col_key, col_print in zip(indices, keys, on_bits):

The variable col_key seems to be unused.

                the_row = []
                if i < j:
                    jaccard = len(row_print.intersection(col_print)) / len(
                        row_print.union(col_print)
                    )
                    the_row.append(jaccard)
                the_rows.append(the_row)
        short = SimilarityDfShortForm(the_rows)
        short["inchikey"] = keys
        short = short.set_index("inchikey")
        short.columns = keys
        return SimilarityDfShortForm.convert(short)


__all__ = ["MatrixPrep"]