Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Ask HN: What problem are you close to solving and how can we help?

news.ycombinator.com

Post reply on HN
Print view View original

111–120 of 486 posts

Re: Ask HN: What problem are you close to solving and how can we help?

#111

Post by Y_Y »

michaelbuckbee wrote: post #97

I could use some help with some heuristics for Machine Learning, like how much data do I need to make a workable model, what framework/approach makes more sense given my ultimate goals. Here's an example: there's a lot of ML tutorials on doing image identification. Like you have a series of images: picture one might have an apple and a pear in it, picture 2 might have an apple, orange, and a banana in it. Where I'm s…

The answer is, "it depends". My recommendation would be to just start trying. What you describe sounds like you could train it in a couple of days. Grab an off-the-shelf resnet and just see what happens. This is a well-studied problem, you can just look at papers that train on imagenet, and then tweak their approaches.

Re: Ask HN: What problem are you close to solving and how can we help?

#112

Post by eitland »

apinstein wrote: post #83
davehcker wrote: post #48

I have posted this here before- hexafarms.com. I am trying to use ML to discover optimal phenotype for growing plants in vertical indoor farms to a. have the higest quality produce b. to lower the cost of producing leafy green/med plants, etc. within cities itself. Basically, every leafy green (and herbs, and even mushrooms), can grow in a range of climatic condition (phenotype, roughly) ie temperature, humidity, wat…

This isn’t an answer to your ML question, but it is an answer to your problem. I heard about a greenhouse company that has programmed their climate control to match “best growing conditions historical weather”. So, they ask local experts what year / location had the best X and then they use that region’s historical weather and replay it in their greenhouse. I thought that was brilliant! (Just realized this was Kimbal…

When I studied farming back in 1998-1999 we once visited a greenhouse and one interesting thing I picked up was that by observation some gardeners had realized that lowering the temperature a bit extra an hour or two before sunrise they could get their flowers to be more compact instead of stretching.

This had replaced shortening hormones in modern gardening (or at least at that greenhouse, but my understanding they were just doing the same thing as everyone else).

I guess there is a lot more to learn for those who have scale enough to experiment and patience to follow through.

Re: Ask HN: What problem are you close to solving and how can we help?

#113

Post by thegginthesky »

michaelbuckbee wrote: post #97

I could use some help with some heuristics for Machine Learning, like how much data do I need to make a workable model, what framework/approach makes more sense given my ultimate goals. Here's an example: there's a lot of ML tutorials on doing image identification. Like you have a series of images: picture one might have an apple and a pear in it, picture 2 might have an apple, orange, and a banana in it. Where I'm s…

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.

Re: Ask HN: What problem are you close to solving and how can we help?

#114

Post by janderson215 »

inglor_cz wrote: post #56

I am a bit ashamed to ask about such a trivial topic on HN, but I am not really sure how AJAX in Wordpress plugins works. I have a plugin that exports some WooCommerce orders into XLS. I would like to add a progress bar via AJAX, because the export may take very long for thousands of orders. But I am not really sure how to use AJAX in context of Wordpress specifically. I would love to see a minimal functional example…

In my experience, accurately reflecting the progress of and AJAX request, so I’ve seen a lot of people (myself included) take the lazy way out and just show an indeterminate spinner or bar just to show that stuff is happening.

Re: Ask HN: What problem are you close to solving and how can we help?

#115

Post by Y_Y »

the-dude wrote: post #6

We are working on a totally new way to do cold fusion, our only problem is getting enough new fuel into the reactor without disturbing the running process. Any help would be greatly appreciated.

What does your reactor look like? What fuel are you using? How are you getting to ignition?

Re: Ask HN: What problem are you close to solving and how can we help?

#116

Post by phenkdo »

yosito wrote: post #63

Ok, I'm double dipping. Another problem I'm trying to solve is: I've got a database of several hundred interesting conversation questions that I've collected over the years. Essentially just strings, though I've attempted to categorize them, rank them, and add other metadata. I'd like to figure out a way to sort them or dedupe them based on semantic similarity, but I'm not sure how to determine semantic similarity wi…

You could use BERT or Word2Vec or GLOVE. They are very simple to use with HuggingFace's library.

Re: Ask HN: What problem are you close to solving and how can we help?

#117

Post by mkl »

tmilard wrote: post #16

https://free-visit.net : Like Matterport but with an FPS game engine. I am looking for my first client: Ideally someone in charge of a Museum/gallery or other grandiose indoor space.

Looks pretty slick. I am not a gamer, and the controls feel very backwards to me. You need to hold the mouse button to look in different directions, which makes it feel like you're dragging, but the direction you move it isn't the direction you're dragging the view. I don't think mouse capture is a good idea in the browser, and if you're aiming at museums and galleries, maybe reversing the mouse direction to make it like familiar dragging would be better.

Edit: There's a bug where if you start dragging with the mouse and let go the mouse button outside the 3D view, it acts like the button is still held down (a bit like mouse capture) which was easier but quite confusing.

It seems unusable with a touch screen on desktop.

The fact that you can fly is useful but non-obvious. I ended up down at floor level and wondering how to see the pictures on the walls.

The demo is in a /fr/ path, but is in English (Chrome offers to translate it to English, because it somehow thinks the English words are French), but then some parts of the interface like "Share your place" are in French.

Re: Ask HN: What problem are you close to solving and how can we help?

#118

Post by nzmsv »

newbamboo wrote: post #39
RHSman2 wrote: post #11

How to get my kids to bed to bed.

Benadryl

Benadryl, no. But melatonin, sometimes, yes. My rule is to have a hard cutoff time after which it's better to take melatonin than to continue the cycle of whining, sleep deprivation, and next-day misery. The cutoff is late enough to have plenty of time to try all the other things involving wind-down rituals. It is not an every day thing. I found that having the consistency actually helps establish the rituals too.

Also, bedtime trouble usually means not enough outside time and physical activity during the day. Or that the kids want more of the parent's time.

Re: Ask HN: What problem are you close to solving and how can we help?

#119

Post by chubot »

These are statistics/math problems that 2 medical professionals I'm seeing are working on, not my own work. But they got me curious. FWIW I worked in "data science" as a software engineer for many years, and did some machine learning, so I have some adjacent experience, but I could use help conceptualizing the problems.

Does anyone know of any books or surveys about statistics and medicine, or specifically mechanics of the human body?

- One therapist is taking measurements of say your arm motion and making inferences about the motion of other muscles. He does is very intuitively but wants to encode the knowledge in software.

- The other one has an oral appliance that has a few parameters that need to be adjusted for different geometries of the mouth and airway.

The problems aren't that well posed which is why I'm looking for pointers to related materials, rather than specific solutions (although any ideas are welcome). I appreciate replies here, and my e-mail is in my profile. I asked a former colleague with a Ph.D. and biostats and he didn't really know. Although I guess biostats is often specifically related to genetics? Or epidemiology?

I guess the other thing this makes me think of is software for Invisalign or Lasik, etc. In fact I remember a former co-worker 10 years ago had actually worked on Lasik control software. If anyone has pointers to knowledge in these areas I'm interested.

Re: Ask HN: What problem are you close to solving and how can we help?

#120

Post by 6510 »

jordwest wrote: post #88

I want to bring back old school distributed forum communities but modernise them in a way that respects attention and isn’t a notification factory. Mastodon is a pretty inspirational project but the Twitter influence shows, I miss the long form writing that was encouraged before our attention spans were eroded. Not at all close to solving it, but it’s been on my mind for a long time. Would love to hear if there are o…

I see a great blog one time where the author was writing a book in the first post. All following postings were dressed up change logs with many interesting and useful comments.

You don't want the notification factory but if its a single post that gets updated you don't get any feed updates at all. Reading the same book again and again looking for the updated sections is also not much fun. Dropping a comment in the long list of comments under it doesn't really create a discussion. (specially not without feed updates)

You could design the publishing tools so that it "forces" the user into that pattern. People could work on multiple books or long reads but start with a crappy draft or just a bunch of links.

Post reply on HN