Hello,
Considering your amazing efficiency on pandas, numpy, and more, it would seem to make sense for your module to work with even bigger data, such as Audio (for example .mp3 and .wav). This is something that would help a lot considering the nature audio (ie. where one of the lowest and most common sampling rates is still 44,100 samples/sec). For a use case, I would consider vaex.open('Hu

Implements {DataFrame,Series}.empty

TurboDBC is probably the fastest method to communicate to a db with arrow data. We can implement utility functions for this in Python.

If turbodbc isn't installed we raise an exception. It probably has to be installed with conda, so we cannot add it as optional dependency.

All required arrow interop functions are already available to make this work.

Describe the bug
Danfojs does not throw correct error when more than one key specified in the merge function

To Reproduce
Steps to reproduce the behavior:

1. Go to 'https://playnotebook.jsdata.org/'
2. load danfojs version 0.2.4 using load_packages
3. run the following code:

let data = [['K0', 'k0', 'A0', 'B0'], ['k0', 'K1', 'A1', 'B1'],
            ['K1', 'K0', 'A

This may just be a problem with data, but I am getting blown up sortino ratios. I thought they had to be much smaller.

For instance:

import yfinance as yf 

msft = yf.Ticker("ETH-USD")

def getSortino(h, p):
  sh = []
  for i in range(0, len(h)):
    if i > p:
      a = h['Close'].iloc[i-p:i]
      sh.append(ta.sortino_ratio(a))
    else: sh.append(0)

  return sh

Hi ,

I am using some basic functions from pyjanitor such as - clean_names() , collapse_levels() in one of my code which I want to productionise.
And there are limitations on the size of the production code base.
Currently ,if I just look at the requirements.txt for just "pyjanitor" , its huge .
I don't think I require all the dependencies in my code.
How can I remove the unnecessary ones ?

I'm using black and isort for other projects (see e.g. https://github.com/hyperopt/hyperopt/pull/748/files) and find them quite useful to have more consistent codebase. I think you should drop python3.5 support though, as black is python3.6+. Is this something you would be open to consider?