<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" ><generator uri="https://jekyllrb.com/" version="3.9.0">Jekyll</generator><link href="https://pratap-ad.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="https://pratap-ad.github.io/" rel="alternate" type="text/html" /><updated>2021-08-24T06:34:18+00:00</updated><id>https://pratap-ad.github.io/feed.xml</id><title type="html">Pratap Adhikari</title><subtitle>DATA SCIENTIST</subtitle><entry><title type="html">Reflection</title><link href="https://pratap-ad.github.io/Reflection/" rel="alternate" type="text/html" title="Reflection" /><published>2020-11-12T00:00:00+00:00</published><updated>2020-11-12T00:00:00+00:00</updated><id>https://pratap-ad.github.io/Reflection</id><content type="html" xml:base="https://pratap-ad.github.io/Reflection/">&lt;p&gt;No major changes about the perspective of data scientist and what they do that are mentioned in the first blog post
 about Data Scientist and Statistician. Still the thoughts about the R is same that it has very sort syntax to execute a very heavy data 
analysis.  So I think it is top on list.&lt;/p&gt;

&lt;p&gt;&lt;br /&gt;
Obviously there is not much doubt that I will be continuing using R going forward. But I have not found a single
job discription that has requirement of having knowledge on R only. There is python or some other languages embeded in the
qualification requirements.&lt;/p&gt;

&lt;p&gt;The things doing differently, One, may be work on other programming languages as well. 
&lt;br /&gt;
Second, I think it would have been very very beneficial if we have got a collaboration with any institutions and work for them on real time data as a project work, 
even voluntarialy.&lt;/p&gt;</content><author><name></name></author><summary type="html">No major changes about the perspective of data scientist and what they do that are mentioned in the first blog post about Data Scientist and Statistician. Still the thoughts about the R is same that it has very sort syntax to execute a very heavy data analysis. So I think it is top on list.</summary></entry><entry><title type="html">Project 2: Create Predictive Models And Automating Markdown Reports</title><link href="https://pratap-ad.github.io/Project-2-Create-Predictive-Models-and-Automating-Markdown-Reports/" rel="alternate" type="text/html" title="Project 2: Create Predictive Models And Automating Markdown Reports" /><published>2020-10-15T00:00:00+00:00</published><updated>2020-10-15T00:00:00+00:00</updated><id>https://pratap-ad.github.io/Project-2:%20-Create-Predictive-Models-and-Automating-Markdown-Reports</id><content type="html" xml:base="https://pratap-ad.github.io/Project-2-Create-Predictive-Models-and-Automating-Markdown-Reports/">&lt;p&gt;The project II is about to automate using R to produce the reports. I have read in and analyzed an online news popularity data set. Shares is the response variable to measure 
the popularity. The model is fitted for everyweek weekdays with their respective subset data. As in the previous project, I think this project has covered all of the major techniques
we have learnt so far. Therefore, I think this is the best approach to get practice to put the education into work to solve the realworld problems. The most difficult part for 
me was in figuring out the use of params to use in the filter function, so that it can be used to reach the variable to compare and retrieve the observations I needed. The most 
important part of this project is the Automoation and understanding how efficient it is for the data scientists who deals with the very big data and reproduction of the similar 
kinds of reports with very few changes in attributes and frequencies of report generation.&lt;/p&gt;

&lt;p&gt;The project work can be found in this &lt;a href=&quot;https://pratap-ad.github.io/Create-Predictive-Models-and-Automating-Markdown-Reports/&quot;&gt;link&lt;/a&gt;&lt;/p&gt;</content><author><name></name></author><summary type="html">The project II is about to automate using R to produce the reports. I have read in and analyzed an online news popularity data set. Shares is the response variable to measure the popularity. The model is fitted for everyweek weekdays with their respective subset data. As in the previous project, I think this project has covered all of the major techniques we have learnt so far. Therefore, I think this is the best approach to get practice to put the education into work to solve the realworld problems. The most difficult part for me was in figuring out the use of params to use in the filter function, so that it can be used to reach the variable to compare and retrieve the observations I needed. The most important part of this project is the Automoation and understanding how efficient it is for the data scientists who deals with the very big data and reproduction of the similar kinds of reports with very few changes in attributes and frequencies of report generation.</summary></entry><entry><title type="html">Project I: Create A Vignette For Reading And Summarizing Data From Nhl Api</title><link href="https://pratap-ad.github.io/Project-I-Create-a-Vignette-for-reading-and-summarizing-data-from-NHL-API/" rel="alternate" type="text/html" title="Project I: Create A Vignette For Reading And Summarizing Data From Nhl Api" /><published>2020-09-18T00:00:00+00:00</published><updated>2020-09-18T00:00:00+00:00</updated><id>https://pratap-ad.github.io/Project-I:%20-Create-a-Vignette-for-reading-and-summarizing-data-from-NHL-API</id><content type="html" xml:base="https://pratap-ad.github.io/Project-I-Create-a-Vignette-for-reading-and-summarizing-data-from-NHL-API/">&lt;p&gt;In this project I created a vignette for reading and summarizing data from the National Hokey League’s (NHL) API. From the pool of data I choosed the franchise ID=15 randomly 
and that came to be of the “Dallas Star”. I tried to analyse the active players in the team, number of games each player played, their wins, losses and ties rates.&lt;/p&gt;

&lt;p&gt;I found most of the players are not ranked as Active players. And the number of recorded wins, losses and ties does not match with the total games played, which was one of the 
discrepencies I found in the records.&lt;/p&gt;

&lt;p&gt;To me personnally, the project approach that we are given is the best way of doing it as it covers the course works we have done so far and we are free to choose any team we like.
In my case, I kow nothing about the game, I can’t say anything more about it if there could be some other prospects of the game that would help modifying the approach.&lt;/p&gt;

&lt;p&gt;In order to get these works done in R, I went throught the given instructions. The first step into the R was installing/loading packages that are necessary to run the code chunks. I was able to write the 
funciton to read the NHL records and stats API. But I was strugginling in writing wrapper function that is essentially a one stop shop for the user to access any of the API endpoints provided. I tried to 
read the data if a user provides a franchiseID#, which was a bit of difficult part of programming to me. Beside that, the biggest chllenging was in Kniting of the document. I don’t know why the R
did knit one time and does not do other time without changing any code. That happened to me on the last assignment as well. I am not sure if it was just a coincidence that I made 
upgrade last week to the latest  R version 4.0.2&lt;/p&gt;

&lt;p&gt;The big-take away is practice more and more in writing R programming functions in order to get familiar with such kind of function which can be encountered in future whether
in the course work and/or in the work life.&lt;/p&gt;

&lt;p&gt;The link to the project is available &lt;a href=&quot;https://pratap-ad.github.io/Create-a-Vignette-for-reading-and-summarizing-data-from-NHL-s-API/&quot;&gt;here&lt;/a&gt;.&lt;/p&gt;</content><author><name></name></author><summary type="html">In this project I created a vignette for reading and summarizing data from the National Hokey League’s (NHL) API. From the pool of data I choosed the franchise ID=15 randomly and that came to be of the “Dallas Star”. I tried to analyse the active players in the team, number of games each player played, their wins, losses and ties rates.</summary></entry><entry><title type="html">Programming Languages</title><link href="https://pratap-ad.github.io/Programming-Languages/" rel="alternate" type="text/html" title="Programming Languages" /><published>2020-08-27T00:00:00+00:00</published><updated>2020-08-27T00:00:00+00:00</updated><id>https://pratap-ad.github.io/Programming-Languages</id><content type="html" xml:base="https://pratap-ad.github.io/Programming-Languages/">&lt;p&gt;The programming languages I am introduced so far are SAS, R,MySql and FORTRAN. I also had tried to self learn Python and SQL programming languages and executed some basic code before I started MS in Statistics in NC-State University, I was certified as a base SAS programmer. During my bachelor’s degree, I had FORTRAN as a course requirement.&lt;/p&gt;

&lt;p&gt;Of course there are pros and cons on all of the programming languages.  I can’t say much about the PYTHON, but what I have percept during the self learning period is that Python is even more advance to R. Comparing to all of programming languages I know. I think R is more user friendly. R does not require any long steps to follow in order to get primary calculations and even on higher level statistical analysis using built in  or using the different packages’ functions. So, I would love to say R is fast, fluent and fun to execute.&lt;/p&gt;</content><author><name></name></author><summary type="html">The programming languages I am introduced so far are SAS, R,MySql and FORTRAN. I also had tried to self learn Python and SQL programming languages and executed some basic code before I started MS in Statistics in NC-State University, I was certified as a base SAS programmer. During my bachelor’s degree, I had FORTRAN as a course requirement.</summary></entry><entry><title type="html">Data Scientist and Statistician!</title><link href="https://pratap-ad.github.io/First-Blog-Post/" rel="alternate" type="text/html" title="Data Scientist and Statistician!" /><published>2020-08-14T00:00:00+00:00</published><updated>2020-08-14T00:00:00+00:00</updated><id>https://pratap-ad.github.io/First-Blog-Post</id><content type="html" xml:base="https://pratap-ad.github.io/First-Blog-Post/">&lt;p&gt;Being a data scientist is about extracting the useful information from data pool with ease focusing/developing machine learning algorithms and being able to tell the data story. 
The data scientists should have knowledge about the statistics, computer science and specific domain to get the insights by ingesting the data. Also, should have the knowledge about data analysis, model selection, model evaluation and communicating the results.&lt;/p&gt;

&lt;p&gt;At least the scale of data dealt by the data scientists and statisticians is kind of incomparable since statisticians still rely on traditional and smaller scale collection method of data like surveys, experiments, polls etc., but they, the data scientists and statistician, are kind of complementary to each other often times at the time of solving the problems. Both play with the data and the purpose of collecting the data is same i.e. to get the useful information but the approaches are different. For example, Data scientists formulate the problems using a modelling process focused on predictive accuracy of the model and it is done by comparing different machine learning algorithms and selecting the best accurate model. Whereas statisticians take different approach to building and testing their model.&lt;/p&gt;

&lt;p&gt;I have dealt only with the structured and small-scale data. Therefore, the knowledge I have so far, refers to be more about the statistician but I believe this Data Science course work will lead towards the data science world with strong statistical background.&lt;/p&gt;</content><author><name></name></author><summary type="html">Being a data scientist is about extracting the useful information from data pool with ease focusing/developing machine learning algorithms and being able to tell the data story. The data scientists should have knowledge about the statistics, computer science and specific domain to get the insights by ingesting the data. Also, should have the knowledge about data analysis, model selection, model evaluation and communicating the results.</summary></entry></feed>