determined VS Plausible Analytics

17. Determined AI | Github | tutorial

Use Determined if you want a nice UI https://github.com/determined-ai/determined#readme

I looked up and found [Determined Platform](determined.ai), tho it looks a very young project that I don't know if it's reliable enough.

- Developer Support Engineer (~1/3 client facing, triaging feature requests and bug reports, etc; 2/3 debugging/troubleshooting)

We are developing enterprise grade artificial intelligence products/services for AI engineering teams and fortune 500 companies and need more software devs to fill the increasing demand.

Find out more at https://determined.ai/. If AI piques your curiosity or you want to interface with highly skilled engineers in the community, apply within (search "determined ai" at careers.hpe.com and drop me a message at asnell AT hpe PERIOD com).

Check out Determined https://github.com/determined-ai/determined to help manage this kind of work at scale: Determined leverages Horovod under the hood, automatically manages cloud resources and can get you up on spot instances, T4's, etc. and will work on your local cluster as well. Gives you additional features like experiment management, scheduling, profiling, model registry, advanced hyperparameter tuning, etc.

Full disclosure: I'm a founder of the project.

These are some of the problems we are trying to solve with the Determined training platform. Determined can be run with or without k8s - the k8s version inherits some of the scheduling problems of k8s, but the non-k8s version uses a custom gang scheduler designed for large scale ML training. Determined offers a priority scheduler that allows smaller jobs to run while being able to schedule a large distributed job whenever you need, by setting a higher priority.

Ah I see - I think we're pretty much on the same page in terms of timetables. Although if you include TPU, I think it's fair to say that custom accelerators are already a moderate success.

Updated my profile. I've been working on DL training platforms and distributed training benchmarking for a bit so I've gotten a nice view into the GPU/TPU battle.

Shameless plug: you should check out the open-source training platform we are building, Determined[1]. One of the goals is to take our hard-earned expertise on training infrastructure and build a tool where people don't need to have that infrastructure expertise. We don't support TPUs, partially because a lack of demand/TPU availability, and partially because our PyTorch TPU experiments were so unimpressive.

[1] GH: https://github.com/determined-ai/determined, Slack: https://join.slack.com/t/determined-community/shared_invite/...

Take a look at Determined https://github.com/determined-ai/determined

determined is an open-source deep learning training platform that makes building models fast and easy. This project provides a CloudFormation template to bootstrap you into AWS and then has a number of tutorials covering how to manage your data, train and then deploy inference endpoints. If you are looking to explore more open source machine learning projects, then check this one out.

I think a single Google Analytics alternative is pretty hard to pick considering that GA can be used to very much varying extents.

For simple and "detailed enough" insights, I enjoyed using Plausible (https://plausible.io/) in the past.

For more in depth analytics that give you a detailed view into your own product, PostHog.com seems to be by far the best and most popular option out there.

I could do the same exercise with Google Analytics and Google Tag Manager, but luckily I don't need to, since Plausible already did. A piece of advice, rip out Google Analytics and use Plausible instead. It first of all doesn't destroy your website, and secondly it doesn't violate the GDPR - So you can embed it on your site without having to warn your visitors about that they're being spied on by Google.

Also, currently we are using https://plausible.io/ for analytics. No other bugs.

I just swapped out Google Analytics with Plausible for AINIRO.IO. It’s only been a week, but so far I am super jazzed about it. First of all, Plausible doesn’t use cookies, so I can completely drop all cookie disclaimers and popups I had because of GDPR. Second of all, the site scores significantly better on load time. This results in a 10x better user experience for my website visitors, while making sure the website is still 100% conforming to GDPR laws.

No clue what you mean, browser cache might even clear itself without you doing anything manually. This thing makes no sense.

Nowhere ever did it say Tech Demo anywhere, not in the HN headline, not on the page itself. No, thanks. And even as a tech demo, there is nothing impressive going in. It is stores shit to local storage, I guess. Lol, I just looked this up, and it was in Firefox on 2009 already? WHAT? https://developer.mozilla.org/en-US/docs/Web/API/Window/loca... I never used it myself directly, but I remember reading about some API that kind of is the new version of cookies that can store more and better and I think that is it. 2009, I would swear what I think about was newer, maybe I am mixing something up, maybe not.

It has unnecessarily tracking from the comment above, not sure if it even sends all your notes to https://plausible.io, and I do not care. For me, this fails as a tech demo or whatever the fuck It's supposed to be. Sorry to not get all excited about everything posted here. In 2009 it for sure would ;)

If you already use Posthog, Web Analytics has been in Public Beta for quite some time.[1]

If I remember correctly, CloudFlare Analytics does not need you to register your domain with them. I personally feel keeping domain registration coupled with your DNS provider is not a good idea.

Plausible[2] has an Open Source self-hostable version but is not so updated in sync with their SaaS version.

Umami[3] is another simple, clean one. And, of course, as many have suggested, Matomo is the other well-established one. If you want to avoid maintaining a hosting routine, a lot do the hosting out of the box these days. PikaPods[4] was good when I tried and played around for a while.

1. https://posthog.com/docs/web-analytics

2. https://github.com/plausible/analytics

3. https://umami.is

4. https://www.pikapods.com

Plausible - Open Source Alternative to Google Analytics

There are many good, lightweight, and open-source alternatives to Google Analytics, such as Plausible, Matomo, Fathom, Simple Analytics, and so on. Many of these options are open-source, and can be self-hosted.

You log the IP address, referrer, user agent and the requested page URL but you don't set a unique cookie to identify the user.

This still gets you plenty of actionable analytics information: where geographically people are located (via GeoIP), what pages are most popular, what platforms (including desktop vs mobile) people are using.

I've been using https://plausible.io for analytics on a bunch of my sites for a couple of years now and I honestly don't miss the extra level of detail I got from cookie-based analytics I've used in the past.