Introduction to Numpy — NCERT Solutions

Use sort(axis=0). In a 2-D array, axis=0 means sorting column-wise, so the lowest value of each column comes in the first row.

myarray1.sort(axis=0)

To split column-wise into 5 arrays, use axis=1 and split at the column indices after each single column. This gives five subarrays:

myarray2A, myarray2B, myarray2C, myarray2D, myarray2E = np.split(myarray2, [1, 2, 3, 4], axis=1)
print(myarray2A, myarray2B, myarray2C, myarray2D, myarray2E)

To split at array indices 2, 5, 7, 8, use np.split() with those indices. This creates four parts:

zerosA, zerosB, zerosC, zerosD = np.split(zeros, [2, 5, 7, 8])
print(zerosA, zerosB, zerosC, zerosD)

To combine the three column parts into a 3 rows and 3 columns array, concatenate them column-wise using axis=1:

np.concatenate((myarray2A, myarray2B, myarray2C), axis=1)

A 2-D array with 14 rows and 3 columns needs $14\times 3=42$ elements. With start value -1 and step size 0.25, use arange() and then reshape:

myarray4 = np.arange(-1, 10.5, 0.25).reshape(14, 3)
parts = np.split(myarray4, 3)
print(parts)

The exact split into 3 equal parts is not possible by rows because 14 is not divisible by 3; however, the chapter method for equal parts is np.split() only when sizes match exactly.

To find the sum of all elements in an array, use the sum() function:

myarray4.sum()

To find the sum row wise, use axis=1:

myarray4.sum(axis=1)

To find the sum column wise, use axis=0:

myarray4.sum(axis=0)

To find the maximum of all elements, use max():

myarray4.max()

To find the minimum in each row, use axis=1:

myarray4.min(axis=1)

To find the mean of all elements in each row, use mean() with axis=1:

myarray4.mean(axis=1)

To find the standard deviation column wise, use std() with axis=0:

myarray4.std(axis=0)

Load the text file into a 2-D array using np.genfromtxt(), skipping the header row, using comma as delimiter, and reading numeric data as float:

iris = np.genfromtxt('C:/NCERT/Iris.txt', skip_header=1, delimiter=',', dtype=float)

To drop the column with index 4, select all rows and only columns 0, 1, 2, 3, 5:

iris = iris[0:30, [0, 1, 2, 3, 5]]

From the chapter’s solution:

iris.shape   # (30, 5)
iris.ndim    # 2
iris.size    # 150

To split the 2-D array into three arrays by species, split row-wise at row indices 10 and 20:

iris1, iris2, iris3 = np.split(iris, [10, 20], axis=0)

Create the 1-D array using np.array() with the given elements in order:

header = np.array(["sepal length", "sepal width", "petal length", "petal width", "Species No"])

Use axis=0 to compute column-wise statistics, and round the mean and standard deviation to at most two decimal places. The chapter’s arrays are:

iris_max = iris.max(axis=0)
iris_min = iris.min(axis=0)
iris_avg = iris.mean(axis=0).round(2)
iris_std = iris.std(axis=0).round(2)

If the question expects iris_var, the chapter does not actually compute variance there; it computes standard deviation and names the array iris_std. So the correct statistical arrays from the chapter are iris_max, iris_min, iris_avg, and iris_std.

Similarly, compute the statistics for each species array using axis=0:

iris1_max = iris1.max(axis=0)
iris1_min = iris1.min(axis=0)
iris1_avg = iris1.mean(axis=0)
iris1_std = iris1.std(axis=0).round(2)

iris2_max = iris2.max(axis=0)
iris2_min = iris2.min(axis=0)
iris2_avg = iris2.mean(axis=0)
iris2_std = iris2.std(axis=0).round(2)

iris3_max = iris3.max(axis=0)
iris3_min = iris3.min(axis=0)
iris3_avg = iris3.mean(axis=0)
iris3_std = iris3.std(axis=0).round(2)

These are the names and results shown in the chapter.

Compare each species minimum with the dataset minimum for the first four columns:

• sepal length: setosa $4.4 > 4.4$ is False
• sepal width: virginica $2.5 > 2.3$ is True
• petal length: versicolor $3.0 > 1.3$ is True
• petal width: setosa $0.1 > 0.1$ is False

So the table entries are False, True, True, False in the order asked.

The chapter gives:

• iris1_avg[1] = 3.31
• iris2_avg[1] = 2.73

Since $3.31 > 2.73$, the comparison is True.

The chapter gives:

• iris1_avg[2] = 1.45
• iris2_avg[2] = 4.03

Since $1.45 > 4.03$ is False, the comparison is False.

The chapter gives:

• iris1_avg[3] = 0.22
• iris2_avg[3] = 1.23

Since $0.22 > 1.23$ is False, the comparison is False.

Use savetxt() with the file name, array, and comma delimiter:

np.savetxt('C:/NCERT/IrisMeanValues.txt', iris_avg, delimiter=',')

To save the three arrays in one text file, use np.savetxt() with a tuple of arrays:

np.savetxt('C:/NCERT/IrisStat.txt', (iris_max, iris_avg, iris_min), delimiter=',')