Numpy square

content/numpy/concepts/math-methods/terms/square/square.md

@@ -0,0 +1,87 @@
+---
+Title: '.square()'
+Description: 'Calculates the square of each element in an array.'
+Subjects:
+  - 'Computer Science'
+  - 'Data Science'
+  - 'Discrete Math'
+Tags:
+  - 'Arrays'
+  - 'Functions'
+  - 'NumPy'
+CatalogContent:
+  - 'learn-python-3'
+  - 'paths/computer-science'
+---
+
+In NumPy, the **`.square()`** method computes the square of a number or the square of the elements in an array. It is commonly used in mathematical calculations in machine learning, data analysis, engineering, and graphics.
+
+## Syntax
+
+```pseudo
+numpy.square(array, out = None, where = True, dtype = None)
+```
+
+- `array`: Input array, it can be a number, an array, or a multidimensional array. Required.
+- `out`: An optional array storing the result. Optional.
+- `where`: (Optional) Used for conditional replacements of the elements in the output array. It must be a numpy
+  array. 
+- `dtype`: The datatype of the output array. Optional.
+
+## Modifying the output array
+
+The output array cannot simply be a Python array since Python arrays are [lists](https://www.codecademy.com/resources/docs/python/built-in-functions/list) of pointers to
+objects. Numpy makes use of arrays composed of contiguous blocks of memory (like in C) for the purposes of optimization. Therefore, arrays supplied for this argument must be initialized with the `numpy.array` function like so...
+
+```py
+output_array = numpy.array([0, 0, 0, 0, 0])
+```
+
+This array can now be called the `out` parameter in your .square() function.
+
+```py
+array = [1, 2, 3, 4, 5]
+numpy.square(array, out = output_array)
+print(output_array)
+```
+
+Output
+
+```shell
+[1, 4, 9, 16, 25]
+```
+
+## Operating conditionally
+
+Using the "where" parameter, the function will execute conditionally. For instance:
+
+```py
+import numpy as np
+array = [1, 2, 3, 4, 5]
+conditions = np.array([False, True, True, False, True])
+np.square(array, where=conditions)
+```
+
+Output
+
+```shell
+array([1, 4, 9, 4, 25])
+```
+
+The "where" parameter takes a boolean value (True, False, 1, 0, etc.) or a matrix of boolean values and squares values at matrix indexes that correspond with true values and vice versa. If the "where" parameter is set equal to a single boolean value, the entire input array is either squared (if it is '1' or 'True') or not squared (if it is '0' or 'False).
+
+## Changing types
+
+Sometimes, it is important to increase or decrease the size of the datatype of the output array. This can be done by setting the 'dtype' parameter to an np datatype, like:
+
+```py
+import numpy as np
+array = [1, 2, 3, 4, 5]
+np.square(array, dtype=np.float32)
+```
+
+Output
+
+```shell
+array([ 1.,  4.,  9., 16., 25.], dtype=float32)
+```