vikp

26 ms·

121.
Predicting HN upvotes from headlines with python(blog.dataquest.io)
10 points by vikp 11y ago|2 comments
122.
by vikp 11y ago
Dataquest (www.dataquest.io) | Boston, MA | Software engineer, data scientist Dataquest is an early stage online education company making "codecademy for data science". Since we launched a few months ago, we've taught tens o
123.
A collaborative machine learning contest(github.com)
2 points by vikp 11y ago|0 comments
124.
by vikp 12y ago
Great change. I second the request for examples -- "gratuitous" is subjective, and until better moderation tools exist, it would be nice if the community has some standards around who to issue "gentle reminders" to. It w
125.
by vikp 12y ago
You can actually create edX classes with existing resources. The xblock spec is pretty flexible, and the existing blocks can be used with a lot of different types of content. There are also a ton of companies that host open edX instances,
126.
by vikp 12y ago
Nice looking site! I worked on something similar a while ago, but it didn't look nearly as good as this. The problem I faced was in attracting users without one key compelling subject area. Saying "this is a platform for every t
127.
by vikp 12y ago
See this issue: https://github.com/jtleek/datasharing/issues/217 .
128.
by vikp 12y ago
I'm also curious about the general case of people believing that we're in a bubble giving advice to startups. Do they advise that startups keep a really low burn rate and wait it out, or go for broke and raise as much as possible
129.
Using naive Bayes to predict movie review sentiment(blog.dataquest.io)
51 points by vikp 12y ago|8 comments
130.
by vikp 12y ago
When I applied, you took the test. If you passed the test, you then had to go online and submit ~10 short essays. If you passed that, you went on to the oral assessments. There may have been a resume component involved when you submitted
131.
by vikp 12y ago
Thanks for sharing! Interesting how we went through a lot of similar steps. I'm glad things worked out well.
132.
by vikp 12y ago
I didn't see this post pre-edit, so I'll assume that it wasn't too bad. If you were suggesting some kind of affirmative action (as someone hinted), then that really wasn't the case. The reason I applied to the Foreign S
133.
by vikp 12y ago
Wow, thanks for sharing. I know how tough of a position to be in this is -- you sound a lot like I was in school. First, dropping out is a serious thing, and I wouldn't take any random internet advice at face value on it. Do you have
134.
by vikp 12y ago
I think the key takeaways that I take from it are (of course, I wrote it, so ymmv): If the system doesn't work for you, it doesn't mean you're not capable. Follow your passions, not external dictates. Try to get as many diffe
135.
by vikp 12y ago
I debated whether or not to write this because, as I mentioned, I have a large degree of impostor syndrome because of my experiences. It's hard to define what "successful" is. The process of "bouncing back" is also
136.
I barely graduated college, and that's okay(vikparuchuri.com)
95 points by vikp 12y ago|60 comments
137.
by vikp 12y ago
I can't help you directly (I'm not in the IT Security field), but I think I have some applicable tips. I did very badly in college, and majored in American History. About three years ago, I quit my job and learned how to code. I
138.
by vikp 12y ago
Using the donor data is a really smart idea. How did you get from donor data to candidate positions? Did you assign manual positions to different kinds of donors and then aggregate them by candidate? I'd love to know more about how
139.
by vikp 12y ago
I made a repo to scrape vote data a while ago: https://github.com/VikParuchuri/political-positions . It's a bit rough, but if you run `scrapy crawl senate -o ../data/senate.json -t json` to scrape, the
140.
Clustering US Senators with k-means(blog.dataquest.io)
87 points by vikp 12y ago|10 comments
141.
by vikp 12y ago
Thanks! Do you know of any organizations that want such a platform? There are a lot of players in the LMS space, like edX, Canvas, etc.
142.
by vikp 12y ago
A few reasons: - Python3 makes some concepts, like unicode, division, the print function, etc, simpler to understand. - Almost all major packages (only scrapy doesn't that I can think of), including the scientific stack, are compatible
143.
by vikp 12y ago
Too long of a list to post here, but I did write a blog post about it a while ago. I need to update it, but these are still relevant: http://www.vikparuchuri.com/blog/resources-for-learning-stat... .
144.
by vikp 12y ago
It's really hard to approach problems from the perspective of a beginner. I did some testing around using "magic" functions like len vs building intuition, and it's really important to understand how the things are work
145.
by vikp 12y ago
Strange, you should be able to click on it to expand. I'll look into it.
146.
by vikp 12y ago
Thanks! The frontend is written in javascript, and uses angular. Design is based on bootstrap. Frontpage is a modified template, rest is custom.
147.
by vikp 12y ago
I actually learned a lot of coding with R. It's hard to learn, but once you do, it's great for hacking up models and trying stuff out. But, because of R's quirks, it's generally easier to write and deploy consistent, go
148.
by vikp 12y ago
Thanks! That's a great idea, and would be good to build out in the medium term. Right now, content is the main bottleneck, and I'd love some help. If anyone is interested in talking, shoot me an email (vik@dataquest.io).
149.
by vikp 12y ago
Hi everyone: Very exciting to see this posted here. I'm the maker of dataquest. I'm a self-taught data scientist/coder, and I wanted an easier way for people to get into the field. I've been working on it for the past t
150.
by vikp 12y ago
I've been working on this for the past 3 months. I'm a self taught coder and data scientist, and I wanted an easier way for people to get into the field. I chose to teach python because it's easier to learn than R, and can b

More ›