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

271–280 of 486 posts

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

#271

Post by yardshop »

cygned wrote: post #38

I try to find an agile project management tool that works for us. We run on what many would call Scrum (it’s not actually Scrum). We are on JIRA now, and it’s … JIRA. We tried basically any other tool, including Excel (yes, that is somewhat possible). My problem generally is that tools are slow, planning is cumbersome, visibility is limited and reporting for clients is often even more limited. Heck, I’d even write my…

I've recently started using ClickUp for managing my helpdesk and development work and I like it a lot. I don't do scrum myself but the product claims to be useful for that kind of work, as well as many other approaches and use cases.

https://clickup.com https://clickup.com/on-demand-demo

ClickUp for Agile Workflows https://www.youtube.com/watch?v=H9hZRwivnL8

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

#272

Post by jpeloquin »

TLDR: What is the state of the art for one-shot or few-shot longitudinal (time-series) machine vision tracking of object boundaries with ~ pixel (~ 10 μm) precision?

Specifics: I'm tracking the edges of the knee meniscus from time lapse video (~ 1000 frames) to measure its deformation under load. This is in the context of research to prevent osteoarthritis. Due to material rotation and irregular geometry, background edges that started off occluded come into and out of view over time. This tends to confuse both machine vision algorithms and human labelers. Because the tracking is for strain measurements, the tracked edge must be the same in all frames; therefore, the tracked edge is be a foreground vs. slightly less foreground edge of the same material (low contrast), not foreground vs. background (much easier). Only few-shot approaches are likely to save time because only ~ 20 specimens are needed to accomplish the immediate objective, and follow-up experiments will probably differ enough to require re-training.

The current plan is to Google "few-shot image segmentation" and try things until something works or the manual labeling effort finishes first, but maybe one of you knows a shortcut. Work is also ongoing to bypass the problem by enhancing edge contrast or using 3D imaging, but machine vision would be the most cost-effective solution.

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

#273

Post by TrackerFF »

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…

Sounds like a problem which YOLO can solve pretty easily , i.e object detection and classification, transfer learning, etc. Try downloading a pre-trained model, and play around with that. (The other replies have outlines what you need to focus on)

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

#274

Post by ConfusedDog »

I’m looking for methods (NLP or otherwise) that maps/generate a short description (few words or less) with a given format (e.g. 3 word tag with each word from a list of choices) from a paragraph or a sentence. Results would be stored and clustered base on their distance

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

#276

Post by Fragoel2 »

udev wrote: post #265

I am blocked on finding a good (defined below) way to determine whether a product description A and product description B refer to the same product. Imagine that a product description is a n-dimensional vector like: ( manufacturerName, modelName, width, height, length, color, ...) Now imagine you have a file with m such vectors (where m is in millions), and that not all fields in the vectors are reliable info (typos,…

Definitely some clustering method based on similarity of the vectors (there are many, pick a simple one to start)

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

#277

Post by tomcooks »

ta988 wrote: post #202
franze wrote: post #122

https://share.securrr.app A secure client side encrypted document (Passport / Id) sharing service Needs a DSVGO lawyer as part of the team.

I love that idea, you should submit that on frontpage.

Frontpage?

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

#278

Post by ethn »

udev wrote: post #265

I am blocked on finding a good (defined below) way to determine whether a product description A and product description B refer to the same product. Imagine that a product description is a n-dimensional vector like: ( manufacturerName, modelName, width, height, length, color, ...) Now imagine you have a file with m such vectors (where m is in millions), and that not all fields in the vectors are reliable info (typos,…

Use a Minhash-LSH ensemble with pre-processing on the words to fix typos via Levenshtein. Tune parameters to get the best distance

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

#279

Post by tomcooks »

29athrowaway wrote: post #199
jeanlucas wrote: post #190

Earlier quoted context omitted.

That's exactly the point, and mostly CTOs don't have much people internally to ask help from. I am in a community with some CTOs in Latin America and I see this struggle happen every time.

Latin America is a mess in many ways. VCs there suck, and do not understand that VCs are about inherently risky investments. They want the guarantees of a low risk business with the profitability of a high risk business. Leadership in Latin American companies also sucks. As soon as the company has any revenue, the leadership will go all out and spend it all on themselves on some extravagant lifestyle adjustments rath…

This applies 1:1 to Italy, I wonder if there's a shared cultural/religious/economic reason. Doesn't seem to apply, for example, to Portugal.

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

#280

Post by cxr »

adriangrigore wrote: post #232

Web site building complexity, for now I started with a simple static site generator https://mkws.sh/ . Right now, I'm not using any package manager, no config files, only one language for templates (sh), and obviously HTML, CSS, Js. I eventually plan to develop a simple CMS based on the same ideas. Ideas and code are welcome!

> (sh), and obviously HTML, CSS, Js

Cut out the sh dependency and just use the "obvious" tech; make the site capable of generating itself, without reliance on any other tools.

Post reply on HN