dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

from .scan_for_files import scan_for_files  # see https://gist.github.com/dmyersturnbull/80845ba9ebab2da83963
from typing import Callable, Iterator

def find_only_file_matching(directory: str, matcher: Callable[[str], bool], file_iterator: Callable[[str], Iterator[str]]=scan_for_files) -> str:
    """Returns the full path of the matching file and raises an exception if none are found or more than 1 is found."""
    file = None

dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

import numpy as np

def centeredness(data: np.ndarray) -> np.float64:
    """A measure of how much the data is concentrated toward the center of a multidimensional array.
    Could probably be defined better. centeredness(arr) == centeredness(arr * 100),
    but centeredness values for arrays of different sizes are not comparable.

dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

from plumbum import local
import os

def make_video_from_frames(input_dir: str, video_path: str, framerate: str='10/1', ffmpeg_path: str='ffmpeg', input_image_extension: str='.jpg') -> None:
    """Runs ffmpeg on all selected files in a directory to generates an x264-encoded video in an MP4 container.
    Warnings:

dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

import pandas as pd
import numpy as np
from typing import Optional

def bootstrap_subtract(X: pd.DataFrame, Y: pd.DataFrame, n_bootstrap_samples: int=200,
        random_seed: Optional[int]=None) -> pd.DataFrame:

dmyersturnbull

import os
from typing import Iterator

def is_proper_file(path: str) -> bool:
    name = os.path.split(path)[1]
    return len(name) > 0 and name[0] not in {'.', '~', '_'}

def scantree(path: str, follow_symlinks: bool=False) -> Iterator[str]:
    """List the full path of every file not beginning with '.', '~', or '_' in a directory, recursively."""
    for entry in os.scandir(path):

dmyersturnbull

from typing import Callable, Generic, TypeVar, Iterable
T = TypeVar('T')

def exists(keep_predicate: Callable[[T], bool], seq: Iterable[T]) -> bool:
    """Efficient existential quantifier for a filter() predicate.
    Returns true iff keep_predicate is true for one or more elements."""
    for e in seq:
        if keep_predicate(e): return True  # short-circuit
    return False

dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

import luigi
from typing import Iterable
import warnings
from .hash_file_utils import *  # seehttps://gist.github.com/dmyersturnbull/f0116c52feae3094f66b0c99b586d166

class AtomicTask(luigi.Task):

dmyersturnbull

# Douglas Myers-Turnbull wrote this for the Kokel Lab, which has released it under the Apache Software License, Version 2.0
# See the license file here: https://gist.github.com/dmyersturnbull/bfa1c3371e7449db553aaa1e7cd3cac1
# The list of copyright owners is unknown

import hashlib
import os
import gzip

algorithm = hashlib.sha1
extension = '.sha1'

dmyersturnbull

import datetime

def format_time(time: datetime.datetime) -> str:
    """Standard timestamp format. Ex: 2016-05-02_22_35_56."""
    return time.strftime("%Y-%m-%d_%H-%M-%S")

def timestamp() -> str:
    """Standard timestamp of time now. Ex: 2016-05-02_22_35_56."""
    return format_time(datetime.datetime.now())

dmyersturnbull

from IPython.display import display, Markdown
import pandas as pd

def head(df: pd.DataFrame, n_rows:int=1) -> None:
    """Pretty-print the head of a Pandas table in a Jupyter notebook and show its dimensions."""
    display(Markdown("**whole table (below):** {} rows × {} columns".format(len(df), len(df.columns))))
    display(df.head(n_rows))