whinvik

17 ms·

181.
by whinvik 3y ago
This is very impressive. Did you use a library for Arrow or did you build your own? We have sometimes faced issues when using the PyArrow library.
182.
Python decorators make composability easier(vikramsg.github.io)
2 points by whinvik 3y ago|0 comments
183.
by whinvik 3y ago
I tried a few profilers for Python a few months back, including Scalene. Memray seemed to be the best of the lot.
184.
by whinvik 3y ago
I didn't quite understand what this meant from the headline. The introduction in the paper clears it up. The TLDR is that astronomers are concerned they won't be able to do space observations with so many satellites in the way, bu
185.
Comparing GPT with Open Source LLM’s(vikramsg.github.io)
2 points by whinvik 3y ago|0 comments
186.
by whinvik 3y ago
> but looking through it it's pretty straightforward, Yes, the code is pretty straightforward. It was mostly going down rabbit holes, or finding API's that was the hard part. > I also find it just exciting to read a post fil
187.
by whinvik 3y ago
Yeah, that would be a cool idea. Will put it on my ToDo list. Nice catch on Heligoland. Thanks. The final journey is via Schiff and I had not filtered that out. However, I don't know if DB categorizes Schiff that are included or exclud
188.
by whinvik 3y ago
> I can choose to be totally spontaneous like take the metro, then local light rail to then take a train to another city and take the metro there then hop on another train to another city, hop off visit a beer garden, hop back on going t
189.
by whinvik 3y ago
Can you expand a bit. What do you mean by "what is a lake or what is a big city"? Thanks for the comment.
190.
by whinvik 3y ago
I had to do a double take. But it is on the WikiVoyage page which gpt decided was worth including in the summary. I will remove it. https://en.wikivoyage.org/wiki/Leipzig
191.
by whinvik 3y ago
That's pretty cool. Do you know if the code is available for this?
192.
by whinvik 3y ago
All of the code is linked to on the website. This is the repo. https://github.com/vikramsg/49travel Or did you mean that I should refer to the code while describing each step?
193.
by whinvik 3y ago
Yes, a number of people have asked for this. This is top of my list of things to do.
194.
Travel across Germany with the 49 Euro ticket(49travel.vercel.app)
188 points by whinvik 3y ago|224 comments
195.
by whinvik 3y ago
Blog here. https://vikramsg.github.io/blog_49travel/
196.
by whinvik 3y ago
It would make sense to write that it is German language only.
197.
by whinvik 3y ago
I feel like this post misses the point. Do most people care if someone was a keynote speaker? No(Outside of academia). Is it a big deal if your talk was selected as a speaker or keynote speaker? No. But should you care if your talk is demot
198.
Numba can be slower than even Pandas, or why you should always benchmark(vikramsg.github.io)
2 points by whinvik 3y ago|0 comments
199.
by whinvik 3y ago
While I think having an ageing audience has something to do with it, I think there's also another influence here. YC and the general startup culture has been to do quick MVP's to validate product. This is great for making companie
200.
Speeding up Pandas GroupBy Apply by using NumPy instead(vikramsg.github.io)
4 points by whinvik 3y ago|0 comments
201.
by whinvik 3y ago
Even though it's not great but I understand why this maybe the direction SciPy maintainers want to go in. They have limited capacity and I think focusing on a single language makes sense especially if there are way more developers in t
202.
by whinvik 3y ago
Going to the Github issue page and reading the exchange is fascinating. Most of the discussion is about licensing and through the thread you see it getting resolved! Also the author's dedication to Prof. Powell seems amazing. I know no
203.
by whinvik 3y ago
It may be a poor analogy but they seem to be doing what marketmakers in the HFT world do for HN up votes!
204.
by whinvik 3y ago
Also interested in the India position.
205.
by whinvik 3y ago
I sent you a mail last month. Never heard back. Would appreciate atleast an acknowledgment/rejection response.
206.
by whinvik 3y ago
I don't know if you are from India but our laws are so complicated that I am sure every moment we are breaking one or the other. If we constantly worried about which ones we are breaking we would never get anything done!
207.
by whinvik 3y ago
I know it's popular to dunk on Facebook and other companies but maybe we should take a moment and think about where Europe is heading. The article mentioned that ad revenue is declining in Europe so maybe that 10% could become 9% very
208.
by whinvik 3y ago
I think you just defined a Startup. The solution is not to do a startup! Other than that, just deal with it?
209.
by whinvik 3y ago
I think it's just about having something new and lots of people getting excited about it. Part of the problem today is that there's too much cynicism floating around especially on sites like Reddit which has seeped into HN also. O
210.
by whinvik 4y ago
I appreciate your response but that is not what I was getting at. I understand that with this you only have to write Pandas and then not worry about scaling. First, I think PySpark syntax is much better than the insanity than is Pandas but

More ›