CS251 OUTLAB 3- Python, Numpy and Scipy Solution

Python, Numpy and Scipy
● Make sure you know what you write, you might be asked to explain your code at a later point in time.
● The submission will be graded automatically, so stick to the naming conventions and follow directory structure strictly.
● In cases where output is required to be submitted in the form of .csv/.txt/.png make sure that your code should generate the file at desired location at runtime.
● After creating your directory, package it into a tarball:rollno1-rollno2.tar.gz
● Command to generate tarball:
tar -czvf rollno1-rollno2.tar.gz rollno1-rollno2/
● Test this on an SL2 machine if you wish.
● Don’t put rubbish like hidden files, temporaries or backup files in the archive.
● Please understand that with great ease you can make grading assignments a nightmare for the TAs. Be responsible and considerate -- this is a two way street.
● Submit once only per team from the moodle account of the smallest roll number.
● The directory structure should be as follows (nothing more nothing less). Quote major references in references.txt.
Directory Structure:
rollno1-rollno2/
|- python
|-q1_a
|-data
|-testcase1
-2024.txt
-olympics.py |-q1_b
|-data
-uid_list.txt
-web_list.txt |-output
-winner.txt (to be created by your program at run time)
-win.py |- numpy
|-q2_a
|-data
-matrix.csv -operations1.py -operations2.py -operations3.py -operations4.py
-operations5.py |-q2_b
|-data
-pca_data.txt |-output
-out.png (to be created by your program at runtime)
-transform.csv (to be created by your program at run time) -pca.py
|- scipy
-q3_a.py
- references.txt
NOTE: All the commands given below considers that terminal has been launched from inside of rollno1-rollno2 directory.
1. Python
CountryName-#Goldmedals-#Silvermedals-#Bronzemedals
.
. .
CountryName-#Goldmedals-#Silvermedals-#Bronzemedals
Read #Goldmedals as the Number of gold medals won by that country,and the same applies for other notations as well. Below given image represents the data from one of the test cases.

Every information that you will read from these files, is going to be in string format. So, please do consider appropriate type conversions. Once you are done with this read part then for every country, you need to store the count of medals separately i.e. for Gold,Silver and Bronze for all the year.txt files at one place.
Now we want you to sort the country names based on their Gold medal count(in decreasing order) and in case of same number of gold medals between two countries tie breaking should be done based on the country names (Just to avoid any sort of confusion between capital letters and small letters , we are providing you the data in small letters only).
For example :
Say, japan won 20 Gold medals and India won 20 Gold medals (Yeah day will definitely come: APNA TIME AYEGA)
Note: Medal count on which sorting needs to be done is the combined sum from all the year.txt files.
Then output should be like : india(should come first) japan(should come second)
i.e. output should be in lexicographic order.In final output, you should output your dictionary as shown in the output of sample run below.
We want you to provide the relative address path through command prompt for better understanding you can look into the following link for how to provide arguments through command prompt Argument_parser_reference.
Sample run:
--2024.txt japan-1-20-50 indonesia-1-30-40 india-1-23-45 China-2-3-4
Command line query to run program: python3 ./python/q1_a/olympics.py --path ./python/q1_a/data/testcase2/
{'usa': [20, 30, 40], 'china': [2, 3, 4], 'india': [2, 25, 49], 'japan': [2, 60, 73],
'indonesia': [1, 30, 40]}
Explanation :
One can clearly see that in the above output usa has the maximum number of gold medals so usa has been ranked first and likewise others but in case of ties like for the case between china, india and japan, they all have equal number of gold medals. So, they have been ranked lexicographically. Try looking for the lambda functions for the implementation. Even new methods are also accepted but do not use any library here other than os,argparse.
Lakshmi Chit Fund is running as one of the most successful online lucky draws. They have got various sponsors from different websites. The sponsors have posted their advertisement links on the Lakshmi Chit Fund website. Now the sponsors want to award prizes to users who click on the entertainment and gaming link.
You will be given a web_list.txt file. This file contains the websites URL that are selected for lucky draws. One URL per line in the format “https://site_path”.
(without double quotes)
(Example : https://www.hotstar.com)
Also you will be given a uid_list.txt file. This file has one entry per line. Each entry has “username” followed by “password” followed by “IP Address” followed by “Website URL” , separated by “||” (double pipe symbol) as shown below :
“user_name||pass_wd||ip_address||url” (without double quotes)
(Example : pranshuiitb||you will never know||10.12.1.2||hotstar.com)
Example:uid_list.txt

Example : web_list.txt

Note : ‘web_list.txt’ is guaranteed to have a complete url starting from https . Example :- “https://www.hotstar.com/entertainment/movies” should print only “www.hotstar.com”.
Example :
print(f“user_name<space>-<space>{user_name}<space>:<space>Winner<sp ace>- Lucky draw!!!<space>-<space>{url_complete})”.
Ex : user_name - Karishma45 : Winner - Lucky draw!!! https://www.allmusic.com/newreleases
Note : In place of <space> you need to replace it with single space.
At last print the number of lucky winners.
Example :

Query : if ‘web_list.txt’ contains https://www.gamespot.com/games/pc , then ‘uid_list.txt’ urls should completely match the later part , i.e. gamespot.com/games/pc is matching with entry in web_list.txt so it should be selected and hence to be printed and copied to ‘winner.txt’ and other entries if any in uid_list.txt like gamespot.com/games should not be printed nor added to winner.txt file.
Task 3 : User code should generate a “winner.txt” inside output directory, which will have one entry per line containing: “user_name||ip_address||complete_url”

Note:
● user_name, pass_w, ip_address, url are separated by || symbol so it will not be present in above attributes in test cases.
● Your program should create a winner.txt file when you run your program. Please make sure that you do not submit winner.txt as a file in your submission.
Command to run win.py file:
python3 ./python/q1_b/win.py --path1 ./python/q1_b/data/web_list.txt --path2 ./python/q1_b/data/uid_list.txt --output ./python/q1_b/output/
2.Numpy
Your friend Rockstar is too good at coding and playing with arrays. He always makes fun of you at coding assessments, since you are a backlogger. You decided to take revenge with your friend. You know that Rockstar only knows C programming language and you are comfortable with python. You know that “Numpy” is a magic wand in python and can work on different functions very easily. So you decided to challenge your friend and one who completes , task will be considered “Coding God”.
Task : You will have a .csv file (matrix.csv) which is a two dimensional matrix, you have to read the csv file and convert it into a numpy array and perform various operations to be coded in these python files named as opertions1.py,opertions2.py….. Operations5.py for the tasks Task1, Task2….Task5 respectively.
Example : Given matrix :
python3 ./numpy/q2_a/operations1.py --path ./numpy/q2_a/data/matrix.csv
Example output :
Example output :
Command to run operations3.py file:
python3 ./numpy/q2_a/operations3.py --path ./numpy/q2_a/data/matrix.csv Example output :

Command to run operations4.py file: python3 ./numpy/q2_a/operations4.py --path ./numpy/q2_a/data/matrix.csv Example output :

Command to run operations4.py file: python3 ./numpy/q2_a/operations5.py --path ./numpy/q2_a/data/matrix.csv
--num 1
Example output :
Note : Use only numpy,pandas packages only. (NumPy,Pandas)
PCA, is a popular statistical dimensionality reduction technique in Machine Learning.It uses an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. Necessary formulas for conversion are given here in the link.
Input : You will be given a txt file (pca_data.txt).Read this file using read_csv() function available in Pandas. This file contains N lines and each line has D comma separated values. Read the .txt file and generate the N x D matrix from it.
Now, for applying PCA on this N x D matrix. You can follow the below given set of steps for PCA :-
Step 1:Standardize the N x D matrix(Do not divide with standard deviation).
Step 2: Calculate the covariance matrix for the D (dimensions) in the matrix .
Step 3: Calculate the eigenvalues and eigenvectors for the covariance matrix.
Step 4: Sort eigenvalues and their corresponding eigenvectors and print sorted eigenvalues on terminal.
Step 5: Pick K=2 (K<D) eigenvalues and form a matrix of eigenvectors.
Step 6: Transform the original matrix and store N x K transformed matrix in the directory “output” with the name ‘transform.csv’
Step 7: Plot the projected data (using a scatter plot from matplotlib) and save the plot to the output directory as ’out.png’ [This saving to file should be automatically done by the code]. While plotting, ensure both the x and y axis have the same aspect, and show values from [-15,15].
NOTE:Your program should save out.png and transform.csv file when you run your program i.e. at run time only. Please make sure that you do not submit out.png and transform.csv as a file in your submission. When we will test your program it should generate these files at the required place.
Command to run pca.py file:
python3 ./numpy/q2_b/pca.py --path ./numpy/q2_b/data/pca_data.txt --output ./numpy/q2_b/output/
Finally you have reduced D dimensions into K (K<D) dimensions, which carry the
same information as D dimensions. Well congratulations you have implemented the PCA algorithm from scratch!!!
Output : Eigenvalues.(Output updated)

Note : Use only os,argparse,numpy,pandas and matplotlib (link).
Do not use scikit or other similar packages for implementing PCA functions. We want you to learn all the intricacies related to numpy library and this is one of the standard examples for learning.
3.Scipy
You all have solved a lot of calculus problems during JEE preparation. Now, Let us get back to scientific calculation but guess what not on paper but on the computer. Yeah, this question will require the use of the scipy library. You can use the following reference link for better understanding of the inbuilt library functions (you can use other resources as well from the web, no restrictions on that but while using them please do put the reference in your reference.txt). We want that you should integrate the following integral:

The answer should be the only thing that you should print on the console screen(i.e. Only the integration result).Use scipy and lambda functions only.
Command to run the program:
python3 ./scipy/q3_a.py 0 0.5 0 3 0 4
NOTE:
● We will check your program on a different set of integral limits but only on this function having triple integral.