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

Show HN: Price Per Ball – Site that sorts golf balls on Amazon by price per ball

priceperball.net

Post reply on HN
Print view View original

31–40 of 52 posts

Re: Show HN: Price Per Ball – Site that sorts golf balls on Amazon by price per ball

#32

Post by rockdiesel »

bombledmonk wrote: post #28

Rather than filter in brands, I'd rather filter out brands (aka exclude). Maybe just an include/exclude toggle there? A select all UI element, then deselect might also be adequate.

Funny that you mention that. That's how it originally worked, but I changed it to filter in brands because I assumed golfers are more brand loyal, and thought they might want to only look at the brands they normally play with.

I think your toggle idea is a good one though, and I'll look to implement that. I can see how some people might want that.

Re: Show HN: Price Per Ball – Site that sorts golf balls on Amazon by price per ball

#34

Post by rockdiesel »

dymk wrote: post #30
rockdiesel wrote: post #12

Earlier quoted context omitted.

Before going full AI, maybe I could create a list of quantity keywords like "dozen", "3-pack", etc... and at least use that as a starting point.

As much as I love simple deterministic things, this is a classic example of where NLP is better than hardcoding a list of keywords. Trying to guess every set of quantity keywords with various spelling, punctuation and how they interact ("1dzn box, two pack" is actually 24 balls) seems more brittle than an LLM.

You're probably right. I can see how LLM is the better way to approach this. I haven't looked into AI usage fees or anything, but I would think the amount of queries I'd send wouldn't be that expensive. I'll look into it.

Re: Show HN: Price Per Ball – Site that sorts golf balls on Amazon by price per ball

#35

Post by fultonn »

rockdiesel wrote: post #7
litenboll wrote: post #5

Great idea, simple and effective. Tiny bit of feedback: seems like some listings use "unit count" for the number of balls, look at the most expensive listing for an example. Annoyingly the second most expensive balls have the number of dozens in the unit count instead.

Yeah. I'm trying to figure out how to combat these inconsistencies. Right now, I have some manual overrides, but not sure it's sustainable to keep manually overriding inconsistent listings. Any thoughts? Should I default to what's in the product title instead of the unit count? Not sure the best way to combat this.

what I've done for a similar script in the past:

    answer_initial = llm(prompt=prompt, site=site) # JSON with answer and any stuff needed to do heuristic checks.
    heuristic_results = heuristics(answer_final) # rule based.
    answer_final = llm(prompt-prompt, site=site, answer=answer_initial)
    mark_for_review = ... # basically just a bunch of hard-coded stuff I add flag possible failures for review.
You can use an extremely small/cheap model for something like this -- granite 4.0 micro works fine for me, 3.3 8b did as well, both run on my macbook. YMMV / try different models and see how it goes.

Re: Show HN: Price Per Ball – Site that sorts golf balls on Amazon by price per ball

#38

Post by parhamn »

rockdiesel wrote: post #8
smokedetector1 wrote: post #4

Somethings broken - the most expensive ball, "TaylorMade 2021 TP5x (3+1 Box) 4DZ" is listed as $169.95 for 1 ball. Clicking through, this is actually for 4 dozen balls, making the actual per ball $3.54

Posting my other comment from above: Yeah. I'm trying to figure out how to combat these inconsistencies. Right now, I have some manual overrides, but not sure it's sustainable to keep manually overriding inconsistent listings. Any thoughts? Should I default to what's in the product title instead of the unit count? Not sure the best way to combat this.

gemini flash!
Post reply on HN