elapid.utils

Backend helper and convenience functions.

check_raster_alignment(raster_paths)

Checks whether the extent, resolution and projection of multiple rasters match exactly.

Parameters:

Name	Type	Description	Default
raster_paths	list	a list of raster covariate paths	required

Returns:

Type	Description
bool	whether all rasters align

Source code in elapid/utils.py

def check_raster_alignment(raster_paths: list) -> bool:
    """Checks whether the extent, resolution and projection of multiple rasters match exactly.

    Args:
        raster_paths: a list of raster covariate paths

    Returns:
        whether all rasters align
    """
    first = raster_paths[0]
    rest = raster_paths[1:]

    with rio.open(first) as src:
        res = src.res
        bounds = src.bounds
        transform = src.transform

    for path in rest:
        with rio.open(path) as src:
            if src.res != res or src.bounds != bounds or src.transform != transform:
                return False

    return True

count_raster_bands(raster_paths)

Returns the total number of bands from a list of rasters.

Parameters:

Name	Type	Description	Default
raster_paths	list	List of raster data file paths.	required

Returns:

Name	Type	Description
n_bands	int	total band count.

Source code in elapid/utils.py

def count_raster_bands(raster_paths: list) -> int:
    """Returns the total number of bands from a list of rasters.

    Args:
        raster_paths: List of raster data file paths.

    Returns:
        n_bands: total band count.
    """
    n_bands = 0
    for path in raster_paths:
        with rio.open(path) as src:
            n_bands += src.count

    return n_bands

create_output_raster_profile(raster_paths, template_idx=0, windowed=True, nodata=None, count=1, compress=None, driver='GTiff', bigtiff=True, dtype='float32')

Gets parameters for windowed reading/writing to output rasters.

Parameters:

Name	Type	Description	Default
raster_paths	list	raster paths of covariates to apply the model to	required
template_idx	int	index of the raster file to use as a template. template_idx=0 sets the first raster as template	0
windowed	bool	perform a block-by-block data read. slower, but reduces memory use	True
nodata	Number	output nodata value	None
count	int	number of bands in the prediction output	1
driver	str	output raster file format (from rasterio.drivers.raster_driver_extensions())	'GTiff'
compress	str	compression type to apply to the output file	None
bigtiff	bool	specify the output file as a bigtiff (for rasters > 2GB)	True
dtype	str	rasterio data type string	'float32'

Returns:

Type	Description
(windows, profile)	an iterable and a dictionary for the window reads and the raster profile

Source code in elapid/utils.py

def create_output_raster_profile(
    raster_paths: list,
    template_idx: int = 0,
    windowed: bool = True,
    nodata: Number = None,
    count: int = 1,
    compress: str = None,
    driver: str = "GTiff",
    bigtiff: bool = True,
    dtype: str = "float32",
) -> Tuple[Iterable, Dict]:
    """Gets parameters for windowed reading/writing to output rasters.

    Args:
        raster_paths: raster paths of covariates to apply the model to
        template_idx: index of the raster file to use as a template. template_idx=0 sets the first raster as template
        windowed: perform a block-by-block data read. slower, but reduces memory use
        nodata: output nodata value
        count: number of bands in the prediction output
        driver: output raster file format (from rasterio.drivers.raster_driver_extensions())
        compress: compression type to apply to the output file
        bigtiff: specify the output file as a bigtiff (for rasters > 2GB)
        dtype: rasterio data type string

    Returns:
        (windows, profile): an iterable and a dictionary for the window reads and the raster profile
    """
    with rio.open(raster_paths[template_idx]) as src:
        if windowed:
            windows = [window for _, window in src.block_windows()]
        else:
            windows = [rio.windows.Window(0, 0, src.width, src.height)]

        dst_profile = src.profile.copy()
        dst_profile.update(
            count=count,
            dtype=dtype,
            nodata=nodata,
            compress=compress,
            driver=driver,
        )
        if bigtiff and driver == "GTiff":
            dst_profile.update(BIGTIFF="YES")

    return windows, dst_profile

download_sample_data(dir, name='ariolimax', quiet=False)

Downloads sample raster and vector files from a web server.

Parameters:

Name	Type	Description	Default
dir	str	the directory to download the data to	required
name	str	the sample dataset to download. options include: "ariolimax" button's banana slug dataset	'ariolimax'
quiet	bool	disable the progress bar	False

Returns:

Type	Description
None	None. Downloads files to dir

Source code in elapid/utils.py

def download_sample_data(dir: str, name: str = "ariolimax", quiet: bool = False) -> None:
    """Downloads sample raster and vector files from a web server.

    Args:
        dir: the directory to download the data to
        name: the sample dataset to download. options include:
            "ariolimax" button's banana slug dataset
        quiet: disable the progress bar

    Returns:
        None. Downloads files to `dir`
    """
    name = str.lower(name)
    https = "https://earth-chris.github.io/images/research"

    if name == "ariolimax":
        fnames = [
            "ariolimax-ca.gpkg",
            "ca-cloudcover-mean.tif",
            "ca-cloudcover-stdv.tif",
            "ca-leafareaindex-mean.tif",
            "ca-leafareaindex-stdv.tif",
            "ca-surfacetemp-mean.tif",
            "ca-surfacetemp-stdv.tif",
        ]

    try:
        os.mkdir(dir)
    except FileExistsError:
        pass

    tqdm = get_tqdm()
    for fname in tqdm(fnames, disable=quiet, **tqdm_opts):
        request.urlretrieve(f"{https}/{fname}", os.path.join(dir, fname))

format_band_labels(raster_paths, labels=None)

Verify the number of labels matches the band count, create labels if none passed.

Parameters:

Name	Type	Description	Default
raster_paths	list	count the total number of bands in these rasters.	required
labels	List[str]	a list of band labels.	None

Returns:

Name	Type	Description
labels		creates default band labels if none are passed.

Source code in elapid/utils.py

def format_band_labels(raster_paths: list, labels: List[str] = None):
    """Verify the number of labels matches the band count, create labels if none passed.

    Args:
        raster_paths: count the total number of bands in these rasters.
        labels: a list of band labels.

    Returns:
        labels: creates default band labels if none are passed.
    """
    n_bands = count_raster_bands(raster_paths)

    if labels is None:
        labels = make_band_labels(n_bands)

    n_labels = len(labels)
    assert n_labels == n_bands, f"number of band labels ({n_labels}) != n_bands ({n_bands})"

    return labels.copy()

get_raster_band_indexes(raster_paths)

Counts the number raster bands to index multi-source, multi-band covariates.

Parameters:

Name	Type	Description	Default
raster_paths	list	a list of raster paths	required

Returns:

Type	Description
(nbands, band_idx)	int and list of the total number of bands and the 0-based start/stop band index for each path

Source code in elapid/utils.py

def get_raster_band_indexes(raster_paths: list) -> Tuple[int, list]:
    """Counts the number raster bands to index multi-source, multi-band covariates.

    Args:
        raster_paths: a list of raster paths

    Returns:
        (nbands, band_idx): int and list of the total number of bands and the 0-based start/stop
            band index for each path
    """
    nbands = 0
    band_idx = [0]
    for i, raster_path in enumerate(raster_paths):
        with rio.open(raster_path) as src:
            nbands += src.count
            band_idx.append(band_idx[i] + src.count)

    return nbands, band_idx

get_tqdm()

Returns a context-appropriate tqdm progress tracking function.

Determines the appropriate tqdm based on the user context, as behavior changes inside/outside of jupyter notebooks.

Returns:

Name	Type	Description
tqdm	Callable	the context-specific tqdm module

Source code in elapid/utils.py

def get_tqdm() -> Callable:
    """Returns a context-appropriate tqdm progress tracking function.

    Determines the appropriate tqdm based on the user context, as
        behavior changes inside/outside of jupyter notebooks.

    Returns:
        tqdm: the context-specific tqdm module
    """
    if in_notebook():
        from tqdm.notebook import tqdm
    else:
        from tqdm import tqdm

    return tqdm

in_notebook()

Evaluate whether the module is currently running in a jupyter notebook.

Source code in elapid/utils.py

def in_notebook() -> bool:
    """Evaluate whether the module is currently running in a jupyter notebook."""
    return "ipykernel" in sys.modules

load_object(path, compressed=True)

Reads a python object into memory that's been saved to disk.

Parameters:

Name	Type	Description	Default
path	str	the file path of the object to load	required
compressed	bool	flag to specify whether the file was compressed prior to saving	True

Returns:

Name	Type	Description
obj	Any	the python object that has been saved (e.g., a MaxentModel() instance)

Source code in elapid/utils.py

def load_object(path: str, compressed: bool = True) -> Any:
    """Reads a python object into memory that's been saved to disk.

    Args:
        path: the file path of the object to load
        compressed: flag to specify whether the file was compressed prior to saving

    Returns:
        obj: the python object that has been saved (e.g., a MaxentModel() instance)
    """
    with open(path, "rb") as f:
        obj = f.read()

    if compressed:
        obj = gzip.decompress(obj)

    return pickle.loads(obj)

load_sample_data(name='ariolimax', drop_geometry=False)

Loads example species presence/background and covariate data.

Parameters:

Name	Type	Description	Default
name	str	the sample dataset to load. options include: "ariolimax" button's banana slug dataset "bradypus" from the R 'maxnet' package	'ariolimax'

Returns:

Type	Description
(x, y)	a tuple of dataframes containing covariate and response data, respectively

Source code in elapid/utils.py

def load_sample_data(name: str = "ariolimax", drop_geometry: bool = False) -> Tuple[pd.DataFrame, pd.DataFrame]:
    """Loads example species presence/background and covariate data.

    Args:
        name: the sample dataset to load. options include:
            "ariolimax" button's banana slug dataset
            "bradypus" from the R 'maxnet' package

    Returns:
        (x, y): a tuple of dataframes containing covariate and response data, respectively
    """
    name = str.lower(name)
    assert name in ["bradypus", "ariolimax"], "Invalid sample data requested"

    package_path = os.path.realpath(__file__)
    package_dir = os.path.dirname(package_path)

    if name == "bradypus":
        file_path = os.path.join(package_dir, "data", "bradypus.csv.gz")
        assert os.path.exists(file_path), "sample data missing from install path."
        df = pd.read_csv(file_path, compression="gzip").astype("int64")
        y = df["presence"].astype("int8")
        x = df.drop(columns=["presence"]).astype({"ecoreg": "category"})
        return x, y

    if name == "ariolimax":
        file_path = os.path.join(package_dir, "data", "ariolimax.gpkg")
        assert os.path.exists(file_path), "sample data missing from install path."
        df = gpd.read_file(file_path)
        columns_to_drop = ["presence"]
        if drop_geometry:
            columns_to_drop.append("geometry")
        x = df.drop(columns=columns_to_drop)
        y = df["presence"].astype("int8")
        return x, y

make_band_labels(n_bands)

Creates a list of band names to assign as dataframe columns.

Parameters:

Name	Type	Description	Default
n_bands	int	total number of raster bands to create labels for.	required

Returns:

Name	Type	Description
labels	list	list of column names.

Source code in elapid/utils.py

def make_band_labels(n_bands: int) -> list:
    """Creates a list of band names to assign as dataframe columns.

    Args:
        n_bands: total number of raster bands to create labels for.

    Returns:
        labels: list of column names.
    """
    n_zeros = n_digits(n_bands)
    labels = ["b{band_number:0{n_zeros}d}".format(band_number=i + 1, n_zeros=n_zeros) for i in range(n_bands)]

    return labels

n_digits(number)

Counts the number of significant integer digits of a number.

Parameters:

Name	Type	Description	Default
number	Number	the number to evaluate.	required

Returns:

Name	Type	Description
order	int	number of digits required to represent a number

Source code in elapid/utils.py

def n_digits(number: Number) -> int:
    """Counts the number of significant integer digits of a number.

    Args:
        number: the number to evaluate.

    Returns:
        order: number of digits required to represent a number
    """
    if number == 0:
        order = 1
    else:
        order = np.floor(np.log10(number)).astype(int) + 1

    return order

repeat_array(x, length=1, axis=0)

Repeats a 1D numpy array along an axis to an arbitrary length

Parameters:

Name	Type	Description	Default
x	array	the n-dimensional array to repeat	required
length	int	the number of times to repeat the array	1
axis	int	the axis along which to repeat the array (valid values include 0 to n+1)	0

Returns:

Type	Description
ndarray	An n+1 dimensional numpy array

Source code in elapid/utils.py

def repeat_array(x: np.array, length: int = 1, axis: int = 0) -> np.ndarray:
    """Repeats a 1D numpy array along an axis to an arbitrary length

    Args:
        x: the n-dimensional array to repeat
        length: the number of times to repeat the array
        axis: the axis along which to repeat the array (valid values include 0 to n+1)

    Returns:
        An n+1 dimensional numpy array
    """
    return np.expand_dims(x, axis=axis).repeat(length, axis=axis)

save_object(obj, path, compress=True)

Writes a python object to disk for later access.

Parameters:

Name	Type	Description	Default
obj	object	a python object or variable to be saved (e.g., a MaxentModel() instance)	required
path	str	the output file path	required

Source code in elapid/utils.py

def save_object(obj: object, path: str, compress: bool = True) -> None:
    """Writes a python object to disk for later access.

    Args:
        obj: a python object or variable to be saved (e.g., a MaxentModel() instance)
        path: the output file path
    """
    obj = pickle.dumps(obj)

    if compress:
        obj = gzip.compress(obj)

    with open(path, "wb") as f:
        f.write(obj)

square_factor(n)

Compute a square form-factor to fit n items.

Parameters:

Name	Type	Description	Default
n	int	the number of items to fit into a square.	required

Returns:

Type	Description
tuple	(x, y) tuple of the square dimensions.

Source code in elapid/utils.py

def square_factor(n: int) -> tuple:
    """Compute a square form-factor to fit `n` items.

    Args:
        n: the number of items to fit into a square.

    Returns:
        (x, y) tuple of the square dimensions.
    """
    val = np.ceil(np.sqrt(n))
    val2 = int(n / val)
    while val2 * val != float(n):
        val -= 1
        val2 = int(n / val)
    return int(val), int(val2)