boondh 0.1.5

A Swiss Knife

boondh

A Personal Library

• Encapsulate boilerplate code
• Data processing code
• ML Frameworks (NLP tasks - SeqClassification, Entity Extractions etc.)

install

pip3 install -U boondh

For nightly

git clone https://github.com/rohit-mehra/boondh.git
cd boondh
pip install .

sample use

Parallelize any function

from boondh import mp_func

# Any function to be applied on each element of data
def base_func(value, sq=True):
    if sq: return value ** 2
    return value

# Any data
data = [0, 1, 2, 3, 4]

# Collected results - Ordered by default
results = mp_func(base_func, # apply this in parallel
                 'value', # the data argument name in above function
                  data, # contains elements to be processed
                  sq=True # any arg of the `to be applied` function
                  )

# configure chunksize and set ordered as False for perfromance gain
results = mp_func(base_func, 'value', data, chunksize=200, ordered=False, sq=True)

Arrange files in your directory (usually the Download directory.. :smile:)

• This will scan the directory and create type folders like csv_, txt_ etc.
• Move respective filetypes into these folders.

python -m boondh.utils.arrange_files -D "$(pwd)"

OR

python -m boondh.utils.arrange_files -D /Users/<myusername>/Downloads/