HNHacker News
TopNewBestAskShowJobs

thegginthesky

532 karma · joined July 2, 2018

Email: gbenatt92 at zohomail dot com
submissionscomments
thegginthesky··on Remote work’s toll on cities
Sure, they had a central plan in the 1800s but only to fit a part of the population that came to live there in the coming centuries. A lot of NYC expansion to support and grow the city was done through displacement[0] and bulldozing[1] of existing buildings.

Most of it wasnt done maliciously, but a city doesn't create square blocks, nice and uniform, on its own as it grows from a few hundred thousand to a few million inhabitants. If it grows organically, as in without a central plan driving it and pruning uncontrolled growth, you will get a very heterogeneous city layout.

[0] https://en.m.wikipedia.org/wiki/San_Juan_Hill,_Manhattan [1]The creative destruction of Manhattan, 1900-1940

thegginthesky··on Remote work’s toll on cities
Yeah, GP has no idea what they are talking about.

I lived with urban planners and architects for 20 years, read books about zoning, about the effects of urbanization on aquifers and rivers, about the differences between cities that grew organically like São Paulo or Mexico City vs cities that were bulldozed to fit a central plan like NYC.

Never ONCE in my life I had read contemporary literature or heard a urban planner defend suburbia, or advocate for single use zoning only, or even say neutral things about American Suburbs. A majority of Urban Planners and Architects across North America, South America and Europe (my sample) are very much against it as it's not sustainable, creates traffic and other myriad of problems.

thegginthesky··on Ask HN: High paying job or better position and comforts
When I moved countries I looked at 3 things

* Quality of life assuming same standard of living * Culture and things to do beyond tourist stuff * If I saw myself in the place for at least 5years

Going to a new country is a huge move, so be 100% sure you will like the city you'll settle and the company you will work for. Once you are there, going back will be financially and emotionally expensive.

Work is work, the pay seems good, just check the true conversion beforehand in sites like this https://neilkakkar.com/salary-calculator-by-city.html

Also make plans to visit your folks regularly if money is no issue.

In regards to the sort of job you'll be performing, maybe you could apply your skills to automate the boring stuff?

Besides that, make sure to be flexible about what you eat and find friends outside work. Invest in building a social circle. I have seen many foreigners that were miserable because they never made real effort to make the new country home and kept missing the old country.

thegginthesky··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
> If you don't mind me asking, what medicine were you prescribed? What kind of topics do you go over in therapy?

First of, I suggest you talking to your doctor about the medicine. For me, I use Vyvanse 30mg, but the same medicine can have a much different impact on you.

I do recommend therapy and really opening up to your therapist. I started out with current struggles, how I felt and handled, and then I talked about past events, formative stuff from my childhood and teenage years.

I also covered topics about inadequacy and frustrations with how I manage my time, my attention and my energy.

Before each sessions I make a list of certain topics I want to cover and I use it as starter, then I let the therapist guide me through some stuff.

thegginthesky··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
> Or figure out a system that works for you. Medication is not the only way.

Exactly, it's good to talk to a professional and see your options. I believe each person can have their own healing mechanism, could be a therapist, exercise or even spiritually related.

I understand everyone is different, made differently with unique physiology, mind and needs.

Sometimes medication can help or harm, for me it's been a positive but not the focus.

thegginthesky··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
Thanks for sharing your experience! I felt and still feel in a similar way.

I don't think negatively of our condition though, it's part of life. I chose to look to change it because of how it impacted my relationship and because I wanted to regain some agency of how I use my time.

But I believe I did fine with it for a number of years before.

I'm glad I sought for help when I did and the short term impacts have been extremely positive for me.

I hope you continue on your path, and I'll be rooting for you.

thegginthesky··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
I searched for a specialist in the area that worked with mainly ADHD, Eating Disorders and so on. In my country, it's easy to find one if you are willing to pay out of pocket, you can also try insurance but the quality of processionals depends on the city you are in.

After I found a few, I was brutally honest with them. I told them about how I truly felt, as for example, when overwhelmed with things around me my mind blanked and I lost complete focus on anything.

I settled with one that had a more balanced view, listened to me, gave me practical advice, and didn't rely on the on just meds or the strongest med.

When it comes to specialists, it pays to search. Although the first consult can be expensive. It's also good to do prior research on their name, my specialist had mostly very positive reviews on many sites, and the reviews didn't seem bots or friends and family.

thegginthesky··on Tell HN: If you have trouble focusing on projects/meetings, see a psychatrist
Interesting!

My doc also told me that everyone is different, and not all medicine works for everyone and most medicine doesn't work at all for some people.

That's why I didn't put emphasis on medicine alone, but I do think it's important to seek for professional help in general. It was important for me, but now I understand it's not a silver bullet.

Thanks a lot for sharing, as it put some new perspective for me and others in this thread.

thegginthesky··on Why it’s hard to buy deodorant in Manhattan
A few blocks away from John Hopkins right? I find that city so bizarre, you have mansions, one of the top universities in the country, and a few blocks away it's pure ghetto. People go from being friendly to staring at you as if you are an alien.

Baltimore is so weird, sometimes feels like living in Rio de Janeiro but only the bad part of Rio. None of the music scene, night life or beaches of Rio, but all of the inequality and violence.

thegginthesky··on GitHub is replacing Rails front end rendering with React
Trur, when I need to browse code I just use codespaces, which you can access by using github.dev/owner/repo, or pressing dot ( . ) when inside the repo
thegginthesky··on Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
Sktime is the best toolkit for time series out there. It provides a sklesrn like API for many models and modules for validating, metrics for evaluating and all that sklearm jazz.

Besides that, I also like statsmodels as the docs are pretty good.

thegginthesky··on Tell HN: Adblock rules to hide HN posts about Twitter & Musk
I, for one, am glad it's not optimized for everyone. It's tough to find places with new exciting content that's adjacent to what I like, which HackerNews does exceedingly well.

I also enjoy the moderation style, which weeds out low-effort posts that plague other communities, so thanks a lot for all the hard work, Dang and team. Keep it up!

Edit: grammar

thegginthesky··on Why U.S. Public Debt Matters [audio]
Exactly, but the goal of a two party system is that you can reliably blame the other party for any problems, even though in practice there's very little difference between both. It's a feature of the current system.
thegginthesky··on The Ethereum merge is done
Another point is that crypto allows for easy transfer of money across borders. Say I want to hire a dev in Germany/Chile/Nigeria, I could go through the cumbersome and expensive process of wiring money, which incurs taxes on money being exchanged, broker fees in the form of a spread and so on, or I could use a cryptocurrency to just send the money and avoid all of that.

Crypto is also a very powerful tool for people in countries with spiraling inflation, such as Venezuela, Turkey and Argentina. Instead of being 100% exposed to local currency or USD, now you have options on how to perform transactions even inside your own country.

thegginthesky··on Show HN: AutoML Python Package for Tabular Data with Automatic Documentation
You do not need to fill missing values or treat categorical with LightGBM.

You can pass the name of the columns that are categoricals when constructing the Booster or Data, and LightGBM will work with them under the hood treating them as a 1 hot encoded.

LightGBM also has a way of automatically treat missing values as either zeros, their own category, or the sample average (I might be mistaken on the last one)

All in all, you still need to do feature engineering and the like, but LightGBM removes a lot of the hassle from Xgboost.

thegginthesky··on Why American cities are broke (2021) [video]
Infrastructure outside the city is normally paid by a combination of County Taxes and State Taxes, and roads are also financed by the Federal Gas Tax, tolls and other forms of government income.

Of course we should pay for infrastructure outside the city, but if you expect to live outside city centers and have costly infrastructure such as a sewage system, high speed fiber internet and so on, you should pay accordingly.

What happens now is that this infrastructure is either financed by those living in city centers or by debt.

So I don't buy your Straw Man argument that people believe that dense urban cores are better. What was explicitly discussed is that people who live in the suburbs don't pay what they should.

thegginthesky··on Ask HN: In 2022, what is the proper way to get into machine/deep learning?
I prefer to use the approach "learn as you go". There's nothing to block someone from learning some basics, a library like Scikit Learn and then learn by doing examples.

But if anyone wants to become an expert into ML/DS, learning statistics and probability is fundamental. Books like A First Course in Probability, Introduction to Statistical Learning and Elements of Statistical Learning, to name a few, are very important.

A lot of the mistakes done in practice are based on a lack of understanding in sampling techniques, how statistical metrics can be misleading and so on.

First thing I learned in statistics is the difference between quantitative and qualitative information. If someone knows this before hopping into Kaggle, they know that categorical features can't be used as continuous features.

thegginthesky··on A billion-dollar industry helping students at Australian universities cheat
I graduated in a country with a free university, and the students still cheated, but most failed. The teachers were more strict with much more challenging assignments and tests, plus a lot of teachers would allow open books, but the difficulty of the exams was out of this world.

Half the class would fail, and because of the difficulty, chances are, If you cheated, it was also wrong.

If the teacher caught you cheating, it was a big fat zero right away, and there would be a board of teachers to protect the decision.

And the University is free,even to re-enroll, you only need to pass the entrance exam, which was super easy for any course except for the doctor, law, or engineering. (CS is super easy to enter because they have more spots than students interested in it)

So I don't know; people will cheat rather than put in the hard work and learn.

thegginthesky··on Bayesian Workflow (2020)
I use it all the time for bayesian testing and hypothesis testing

A few resources

https://docs.pymc.io/en/v3/pymc-examples/examples/case_studi...

https://docs.pymc.io/en/v3/pymc-examples/examples/case_studi...

thegginthesky··on Ask HN: Jobs After a Math PhD?
Many Data Science/Data Analyst jobs would love to have someone with your credentials. Also, if you self-teach how to code and create software, a math PhD can be a huge boost for certain specializations (i.e. Machine Learning Engineer)
thegginthesky··on Anatomy of the Presidential Motorcade (2016)
Got it, but I think most democracies (for example US, Brazil, Argentina) have a line of succession either in their constitution or in the amendments that follow it. Whenever anyone takes the role of president next, they have full authority, power, and responsibilities of a president, even if in some cases they don't have the title of President.

I know of a few examples that it's required to have new elections months after a president step down, which is the case of India, but even then the acting president has the full power of a president.

thegginthesky··on Anatomy of the Presidential Motorcade (2016)
Why would it take months for a new president? In case of the death of a president, the vice-president gets automatically sworn into office as fast as possible and in fact becomes a president. Now if the vice president also dies, you have a line of succession to act as a president.

I haven't found any information about the obligation to hold new elections in such event, but I could be wrong

thegginthesky··on What software engineers can learn from the rapid collapse of Fast
Agreed.

I've been on companies that did the opposite from going small to big. They started big, very big, with Fortune 50 costumers. That meant they hired a small but very specialized sales team with the right connections very early on.

Because the burn rate was very small (due to needing just a few engineers and employees and a small infrastructure) and the sales cycle took months, we could easily develop a strong product with the requested features, and then it turned out when we moved down to medium and small businesses most of the features were already in place all we needed to focus was on scaling the architecture.

Plus, big costumers pay very good and are loyal as long as you don't cause them headaches.

The only downside of this approach is if we hired the wrong sales people, the company would never take off.

thegginthesky··on Lawn mowing frequency affects bee abundance and diversity (2018)
Depends on the location. Southern states in the US and many countries in Latin America have strict rules against rubble that accumulates still water, and you can even get fined if you have a pool and don't keep it clean and treated.

West Nile virus and Dengue virus are serious health hazards transmitted by mosquitoes, and many places try to contain it.

thegginthesky··on Historical Redlining and Present-Day Air Pollution Disparities in U.S. Cities
I should've read the paper carefully, because I found a link for it in the references

https://dsl.richmond.edu/panorama/redlining/#loc=6/39.741/-7...

thegginthesky··on Historical Redlining and Present-Day Air Pollution Disparities in U.S. Cities
Does anyone know of a good dataset on redlining? I'm looking for the polygon coordinates and if it was red, yellow, green or blue. I found several maps in image/pdf format but nothing that I could easily translate to Zip Code Tabulation Área.
thegginthesky··on Launch HN: Slai (YC W22) – Build ML models quickly and deploy them as apps
Congrats on the launch!

Overall I like the idea and I agree with you, either the tools are too focused on Data Scientists or there are a lot of DevOps involved to get things started

I work on the field so I have some questions:

- Are there any plans to connect the project into a git repo?

- Is there any option for me to pass trained binaries to your product? For example I have a beast of a machine and I can easily train things locally, but I'd like to host the inference with you guys

- Do you intend to allow automated testing and linting?

thegginthesky··on U.S. companies are hiring Latin America’s tech talent
This is a misconception that a few hundred of devs in a city can do this.

What drives the price of real state up is lack of new developments, low interest rates, and instutional investors using the housing market as a security instrument. The same thing can happen all over the world and is happening in the US and Canada at an alarming rate.

thegginthesky··on It's hard to say who's winning the streaming wars, but customers are losing
The biggest difference is that people use the internet for far more things than just streaming, so the internet bill is diluted into the overall utility. For example, without good internet speeds you couldn't attend classes properly in lock down, nor attend meetings with people across the country or world.

Good internet is becoming more of an overall utility than just a luxury.

thegginthesky··on Ask HN: What problem are you close to solving and how can we help?
Hey, I'm a ML practitioner for over 6 years and I'm glad to help.

1k distinct labels with a long tail distributions which is what you are describing, is definitely a challenging problem. It's called a imbalanced classification problem.

I'd focus first test how well your model will be able to predict these classes doing a stratified cross validation (stratification controlled by the label class), and measure the F-score, Weighted Accuracy and ROC AUC. Check also the precision and recall for each class. You'll definitely see that the model predicts better for the labels with more samples. The code that you use here, you'll be able to reuse later on, so keep it organized and easy to follow.

Then you have a couple options, focus on gathering more examples for the labels with small sample size, or try to oversample your dataset. This article is a good place to start https://towardsdatascience.com/4-ways-to-improve-class-imbal...

Considering this problem of image classification is normally solved with deep learning, the more data you have, the better will be your results.

← PreviousPage 2 of 4Next →