| from datetime import datetime | |
| import subprocess | |
| import json | |
| import requests | |
| class SlackNotifier: | |
| def __init__(self, slack_url, msg='Done!'): | |
| self.start = datetime.now() |
| import pandas as pd | |
| import numpy as np | |
| df = pd.DataFrame([[1.,2.,np.nan,np.inf],[1.,2.,3.,4.,]]).T | |
| df[np.isnan(df).any(axis=1) | np.isinf(df).any(axis=1)] |
| # Check nthreads() is large enough | |
| println(Threads.nthreads()) | |
| function f(input) | |
| return input*2 | |
| end | |
| function main() | |
| inputs = [i for i in 20:30]; |
Here's how to transfer your data frames between Python, R, and Julia by using Feather format. An article The Best Format to Save Pandas Data provides speed comparisons with other formats and concludes Feather is the best format to store data for a short term.
Note for pandas users: Feather doesn't support data frames with a custom index. Execute df.reset_index() before writing your data in a Feather file if you store anything in the pandas index.
import pandas as pd
import feather
# read
| from multiprocessing import Pool | |
| def func(a): # in case you want to pass multiple arguments, pack them by making a data type, or use starmap | |
| b = a**2 | |
| return a, b # return the argument in addition to the result for convenience | |
| with Pool(20) as p: | |
| output = {k: v for k, v in p.imap(func, range(100))} |
| #!/bin/sh | |
| for i in $(seq 01 10); do | |
| fallocate -l 1GB dummy${i} | |
| done | |
| for i in $(seq 01 10); do | |
| echo "File${i} start, $(date +%T)" | |
| cp dummy${i} /foo/bar/destination/ | |
| done |
| def get_colorcode(colormap, value): | |
| """ | |
| Gets hex format colorcode (eg. '#ff0000') from a matplotlib colormap. | |
| Parameters | |
| ---------- | |
| colormap : matplotlib.colors.Colormap | |
| Matplotlib colormap, a subclass of Colormap | |
| value : float | |
| Value in [0, 1) interval that specifies the color |
| import numpy as np | |
| import pandas as pd | |
| from ete3 import PhyloTree | |
| from itertools import cycle | |
| def generate_colorstrip(labels: pd.Series, save_file: str, palette: list = None) -> None: | |
| # labels must be a pd.Series whose index are leaf names and values are their labels | |
| if palette is None: | |
| palette = ["#4E79A7", "#A0CBE8", "#F28E2B", "#FFBE7D", "#59A14F", | |
| "#8CD17D", "#B6992D", "#F1CE63", "#499894", "#86BDB6", |
| def prime_factorize(n): | |
| answers = [] | |
| while n % 2 == 0: | |
| answers.append(2) | |
| n //= 2 | |
| f = 3 | |
| while f**2 <= n: # instead of f <= n**0.5 | |
| if n % f == 0: | |
| answers.append(f) | |
| n //= f |
| def get_gcd(a, b): | |
| if b > a: | |
| a, b = b, a | |
| r = a%b | |
| if r == 0: | |
| return b | |
| else: | |
| return get_gcd(b, r) |