Assignment - agastya2002/IECSE-ML-Winter-2020 GitHub Wiki
Assignments Week 1
Deadline 18 Jan 2021
Assignment 1 - Basic Python
Important
Before you turn this problem in, make sure everything runs as expected. First, restart the kernel (in the menubar, select Kernel -> Restart) and then run all cells (in the menubar, select Cell -> Run All).
Make sure you fill in any place that says YOUR CODE HERE or "YOUR ANSWER HERE"
Note: For the questions given below, we have given you along with it examples of each question, and also the basic template of the function. we have also provided the test cases that need to be passed in the form of assert statements. You can look up assert statements in python online. If your assert statements don't give any error that means your code works fine and has satisfied the test cases. You just need to copy paste the assert functions at the end of your code to check if your program function works. It is applicable for both native python installations as well as jupyter-notebooks.
Note: You guys may may search online for certain functions, but we want you guys to try your best while attempting these assignments as they are fundamental in improving your grasp over Python. You can always discuss solutions with your friends, but copying your friend's code is discouraged. Rather we encourage you to ask doubts on the group or else approach your mentors for any difficulties.
1
Find the last element of a list. Eg., l = [1, 2, 3, 4] last_element(l) -> 4
def last_element(l):
"""
Returns last element of list l
"""
# YOUR CODE HERE
Your function should return 4. Check that it does
last_element([1, 2, 3, 4])
4
"""Testing code for last_element"""
assert last_element([1]) == 1
assert last_element(["Hello", "World"]) == "World"
2
Find the number of elements in a list. Eg., l = [1, 2, 3, 4] num_elements(l) -> 4
def num_elements(l):
"""
Returns number of elements in the list l
"""
# YOUR CODE HERE
"""Testing code for num_elements"""
assert num_elements([2, 3, 4])==3
3
Reverse a list and return it. Eg., l = [1, 2, 3, 4] reverse_list(l) -> [4, 3, 2, 1]
def reverse_list(l):
"""
Returns the reverse of a list l
"""
# YOUR CODE HERE
"""Testing code for reverse_list"""
assert reverse_list([2, 3, 4])== [4, 3, 2]
4
Find whether a list a palindrome -> a sequence that reads the same forwards and backwards is_palindrome([1, 2, 1]) -> True is_palindrome([1, 2, 3, 2, 1] -> True is_palindrome([1, 2, 3, 1]) -> False
def is_palindrome(l):
"""
Returns True if l is a palindrome, False otherwise
"""
# YOUR CODE HERE
"""Testing code for is_palindrome"""
assert(is_palindrome([1, 2, 1])) == True
assert(is_palindrome([1, 2, 3, 2, 1])) == True
assert(is_palindrome([1, 2, 3, 4])) == False
5
Replace consecutive duplicate elements of list with single element. Eg., l = [a, a, a, b, b, c, a, a, d, d, d, x, x] compress(l) -> [a, b, c, a, d, x]
def compress(l):
"""
Returns a list with consecutive duplicate elements replaced by a single element
"""
# YOUR CODE HERE
"""Testing code for compress"""
assert(compress([1, 2, 2])) == [1, 2]
assert(compress([1, 2, 2, 2, 1, 1, 3, 'x', 'x', 'x'])) == [1, 2, 1, 3, 'x']
6
Pack consecutive duplicates of list elements into sublists. Eg., l = [1, 1, 1, 2, 2, 3, 3, 4] pack(l) -> [ [1,1,1], [2,2], [3,3], [4] ]
def pack(l):
"""
Returns a list with consecutive duplicate elements packed into sublists
"""
# YOUR CODE HERE
"""Testing code for pack"""
assert(pack([1, 1, 1, 2]) == [1, 1, 1], [2](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-1,-1],-[2))
assert(pack([1, 1, 1, 2, 1, 1, 3, 3, 3])) == [1, 1, 1], [2], [1, 1], [3, 3, 3](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-1,-1],-[2],-[1,-1],-[3,-3,-3)
7
Given two indices, i and k, the slice is the list containing the elements between the i'th and k'th element of the original list (left limit included). Start counting the elements with 0.
Eg., l = [1, 3, 9, 8, 7]; slice(l, 1, 3) --> [3, 9]
def slice(l, i, k):
"""
Returns a list containing the elements between i'th and k'th elements of original list l.
"""
# YOUR CODE HERE
"""Testing code for slice"""
assert(slice([1, 3, 8, 9, 7], 1, 3)) == [3,8]
assert(slice([1, 4, 6, 'x', 9, 0], 2, 10)) == [6, 'x', 9, 0]
8
Given a list l, index i and element elem, return a new list with elem at index i Eg., l = [1, 3, 9, 8, 7]; insert_element(l, 1, 3) --> [1, 3, 3, 9, 8, 7]
def insert_element(l, i, elem):
"""
Returns a new list containing elem at index i. If i > len (l), insert element at the end of the list
"""
# YOUR CODE HERE
"""Testing code for insert_element"""
assert(insert_element([1, 2, 3, 4,], 2, 5)[2]) == 5
assert(insert_element([1, 5, ], 3, 5)) == [1, 5, 5]
Assignment 2 - NumPy Basics
Instructions :
- Only Edit YOUR CODE HERE
- Do not change the assert statements which test your functions
import numpy as np
Windows and Strides
Given a 1d array, and parameters window_len and stride_len, do the following:
- Divide the 1d array into smaller arrays of size window_len starting at index 0
- move the starting index by step size stride_len each time
- Return all windows as a 2d matrix
Eg., Input: arr = [1, 3, 7, 1, 2, 6, 0, 1] stride_len = 2 , window_len = 4
- First window (starting at index 0) should be of len window_len (4) --> [1, 3, 7, 1]
- Move by step size stride_len(2) and take next window --> [7, 1, 2, 6] and so on..
Output matrix [[1, 3, 7, 1], [7, 1, 2, 6], [2, 6, 0, 1]]
def gen_strides(a, stride_len, window_len):
'''
Input:
a: Numpy array of 1 dimension
stride_len: int, stride length
window_len : int, window length
Output:
Numpy array of 2 dimensions containing windowed strides as explained above
'''
# YOUR CODE HERE
"""Test for strides"""
assert (np.all(gen_strides(np.array([1, 3, 7, 1, 2, 6, 0, 1]),2,4) == np.array([1, 3, 7, 1], [7, 1, 2, 6], [2, 6, 0, 1](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-3,-7,-1],-[7,-1,-2,-6],-[2,-6,-0,-1))))
print("Sample Tests passed", '\U0001F44D')
Shuffle
Given a numpy array of arbitrary dimensions (> 1), shuffle its rows randomly. Hint - You need to shuffle along axis 0
def shuf(arr):
'''
Input:
arr: Numpy array of arbitrary number of dimensions (>1)
Output:
numpy array of same shape as arr but with rows shuffled
'''
# YOUR CODE HERE
"""Test for shuf"""
arr=np.array([1, 2, 3],[4, 5, 6],[7, 8, 9](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2,-3],[4,-5,-6],[7,-8,-9))
assert np.any(shuf(np.array([1, 2, 3],[4, 5, 6],[7, 8, 9](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2,-3],[4,-5,-6],[7,-8,-9))) != np.array([1, 2, 3], [4, 5, 6], [7, 8, 9](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2,-3],-[4,-5,-6],-[7,-8,-9)))
assert shuf(np.array([1, 2, 3],[4, 5, 6],[7, 8, 9](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2,-3],[4,-5,-6],[7,-8,-9))).shape == np.array([1, 2, 3], [4, 5, 6], [7, 8, 9](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2,-3],-[4,-5,-6],-[7,-8,-9)).shape
print("Sample Tests passed", '\U0001F44D')
Match
Get the positions where corresponding elements (same indices) of array a and array b match
Eg. Input <br>
a = np.array([1,2,3,2,3,4,3,4,5,6])
b = np.array([7,2,10,2,7,4,9,4,9,8])
Desired Output:
[1, 3, 5, 7])
def match(a,b):
'''
Inputs:
a, b: numpy arrays of same shape of 1 dimension
Outputs:
list containing indices where both arrays have same elements
'''
# YOUR CODE HERE
"""Test for match"""
assert(match(np.array([1,2,3,2,3,4,3,4,5,6]),np.array([7,2,10,2,7,4,9,4,9,8])) == [1,3,5,7])
print("Sample Tests passed", '\U0001F44D')
Inverse of an array
Hint: Search numpy library for inverse function
def inv(arr):
"""
Given an array arr (square matrix), find its inverse
"""
# YOUR CODE HERE
"""Test for inv"""
assert np.all(np.isclose(inv(np.array([6, 1, 1], [4, -2, 5], [2, 8, 7](/agastya2002/IECSE-ML-Winter-2020/wiki/6,-1,-1],-[4,--2,-5],-[2,-8,-7))).tolist(), np.array([0.17647058823529413, -0.0032679738562091526, -0.02287581699346405],[0.05882352941176469, -0.130718954248366, 0.0849673202614379],[-0.1176470588235294, 0.1503267973856209, 0.0522875816993464](/agastya2002/IECSE-ML-Winter-2020/wiki/0.17647058823529413,--0.0032679738562091526,--0.02287581699346405],[0.05882352941176469,--0.130718954248366,-0.0849673202614379],[-0.1176470588235294,-0.1503267973856209,-0.0522875816993464))))
print("Sample Tests passed", '\U0001F44D')
Linear Equations
Hint: x = inverse_of_a * b.
def lin_eqn(a,b):
'''
Solve the system of linear equations
of the form ax = b
Eg.
Solve the system of linear equation
x + 2*y = 8
3*x + 4*y = 18
Given inputs a and b represent coefficients and constant of linear equation respectively
coefficients:
a = np.array([1, 2], [3, 4](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2],-[3,-4))
constants:
b = np.array([8, 18])
Desired Output: [2,3]
'''
# YOUR CODE HERE
x=lin_eqn(np.array([1.0, 2.0], [3, 4](/agastya2002/IECSE-ML-Winter-2020/wiki/1.0,-2.0],-[3,-4)),np.array([8.0, 18.0]))
print(x)
[2. 3.]
"""Test for lin_eqn"""
assert np.any(lin_eqn(np.array([1, 2], [3, 4](/agastya2002/IECSE-ML-Winter-2020/wiki/1,-2],-[3,-4)),np.array([8, 18])) == np.array([2., 3.]))
print("Sample Tests passed", '\U0001F44D')
rankArray
Rank the items in a multidimensional array 'arr'. The rank of an item is its index in the sorted list of all items in 'arr' (starting from 0). Eg., consider arr = [0, 6, 14, 12, 11] arr_sorted = [0, 6, 11, 12, 14] So rank of 0 -> 0; 6->1; 11->2, 12->3, 14->4 rankArray returns a list where each element is replaced by its rank rankArray(arr) -> [0, 1, 4, 3, 2]
==========================================
Another example,
Eg:
arr = [ [ 9 4 15 0 17], [16 17 8 9 0] ] ... (2X5 array)
Desired output: [ [4 2 6 0 8], [7 9 3 5 1] ] ... (2X5 array)
Here minimum value in arr is 0 at (0,3), so rank the index of 0 in output matrix as 0 i.e. out[0][3] = 0 Next minimum is also 0 at index (1,4), so rank the index (1,4) in output matrix as 1 and so on...
============================================= If two elements repeat; for eg., both a1 = arr[0, 1] and a2 = arr[0, 3] are equal to 1, then rank of the a1 is lower than the rank of a2 because at the first index at which they differ, the index of a1 is lower than that of a2 (similar to lexicographical ordering)
import numpy as np
def rankArray(arr):
'''
Input:
arr: Numpy array of arbitrary dimensions
Output:
numpy array of same shape as arr but with elements replaced by their ranks
'''
# YOUR CODE HERE
print(rankArray(np.array([9, 4, 15, 0, 17], [16,17,8,9,0](/agastya2002/IECSE-ML-Winter-2020/wiki/9,-4,-15,-0,-17],-[16,17,8,9,0))).tolist())
[4, 2, 6, 0, 8], [7, 9, 3, 5, 1](/agastya2002/IECSE-ML-Winter-2020/wiki/4,-2,-6,-0,-8],-[7,-9,-3,-5,-1)
"""Test for rankArray"""
assert np.all(rankArray(np.array([9, 4, 15, 0, 17], [16,17,8,9,0](/agastya2002/IECSE-ML-Winter-2020/wiki/9,-4,-15,-0,-17],-[16,17,8,9,0))) == np.array([4,2, 6, 0, 8], [7, 9, 3, 5, 1](/agastya2002/IECSE-ML-Winter-2020/wiki/4,2,-6,-0,-8],-[7,-9,-3,-5,-1)).tolist())
print("Sample Tests passed", '\U0001F44D')
Assignment 3
Plot the data given in the dataset, the link to which you shall find here. Your plot should look something like this. Push your code and also upload a screenshot of the plot on your branch.
Assignment 4
Watch this video by Andrew Ng to get a clear understanding of what machine learning is. You will find the link to the video here.
Note for WC
Solutions to assignment questions should be uploaded on the repo on your own branch. Do not push them to main.