NumPy: Python Library for Data Science
Importing the library
import numpy as np
Check version
np.__version__
Create NumPy array from python list
arr = np.array([1,2,3,4])
Check type
type(arr)
# Output : <class 'numpy.ndarray'>
# arr is n-dimensional numpy array
Check datatype of elements in numpy array
arr.dtype
# Output : int64
Check dimension of numpy array
arr.ndim
# Output : 1
# arr is 1-dimensional arraynp.array([[1,2],[3,4],[5,6]]).ndim
# Output : 2
# arr is 2-dimensional array
Check shape of numpy array
Changing the shape of numpy array
Reshape: Changing the shape of numpy array
Create an array in a range using ‘arange’
default np.arange(start=0, stop, step=1, dtype=‘int’)
‘linspace’
Returns evenly spaced samples over the interval [start, stop]
np.linspace(start=0, stop, nums=50, endpoint=True, retstep=False, dtype=None)
Statistics
Transpose
‘np.where’
‘argmin’ works with numpy array or lists
‘idxmin’ works only dataframes and series
drop_duplicates : Return DataFrame with duplicate rows removed.
To learn more about NumPy, check out the official documentation here.
In next blog we’ll discuss about another important library of python- ‘Pandas’ which makes working with dataset easy and is extensively use in Data Science.
Happy Learning!!