Greetings! I'm doing a research on Plautdietsch-speaking Mennonite communities in South America, however, while written Plautdietsch is having a revival in these communities, the majority of them seem to only use it in oral communication, while the rest of their literature and newspapers are in standard German, so I think that the latter will be more useful for this study.
I was learning standard German as a hobby until around 2014 or so when I lost interest, I thought that I would fare well in only learning Plautdietsch from scratch without knowing too much of standard German, but I do not think that this will be too helpful, so would it be better for me to fully immerse myself into standard Deutsch first before switching over to Plautdietsch?
Another issue is that there is very few online learning material for Plautdietsch (for obvious reasons, given that its speakers are almost all rural Mennonites who use few technology), other than a few not-so-helpful youtube videos, the other sources that I found are like 3 old Plautdietsch dictionary sites and a translation of the New Testament that is freely available.
Also again, this is about Plautdietsch-speaking Russian Mennonite communities in South America like in Paraguay and Bolivia, not those from Canada and the US who are more numerous, so most of these Plautdietsch speakers I am researching and eventually entering contact with do not speak with an American accent and do not have too much influences from English, which would've made it easier for me.
Worth mentioning that (Brazilian) Portuguese is my native language, but I also speak English, Spanish and Italian, so since English is the only Germanic language I speak with fluency, up until recently I was trying to learn Plautdietsch by trying to understand it from the lens of an English speaker, as I had no intentions of learning standard Deutsch, hence why I think that I am beginning to learn it in a wrong way.
Hi all,
Merry Christmas to you all. I come for a bit of wisdom from you.
What are some fundamentals that would be solid foundations on which someone could start learning DIY for things to do around a standard home?
If there are any particular videos or guides that teach these things it would be even better.
I would like to do more DIY around my home in 2022 and onwards so thought to ask.
As the title states, I've grown tired of the endless numbers of superficial resources on the internet for learning Machine Learning. Over the last couple of years, I've been fortunate enough to be exposed to some excellent teachers so that by now, I've accumulated a decent wealth of knowledge and understanding of Machine Learning and Statistical Learning Theory in particular. And since I'm passionate about education, I want to give anyone access to these deep insights into how we and "artificial systems" can make sense of data. So I started a YouTube channel and I've already created my first few videos: https://www.youtube.com/channel/UCg5yxN5N4Yup9dP_uN69vEQ
The feedback so far has been great and this really motivates me to keep going. This first playlist will be an 8 lectures long series on Regression and Kernel Methods. We start tame with some simple prerequisites but by the end we will have covered the Reproducing Kernel Hilbert Spaces and their Mercer representation, before concluding with their not-at-all obvious relationship to Gaussian process regression which will bridge the gap between the frequentist interpretation of the kernel formalism and the Bayesian framework of evidence based belief updates.
Future playlists will follow, where I'll cover even more advanced topics like Geometric Deep Learning (which is a unifying formalism for all of Deep Learning and finally provides some rigorous statements of why some NN architectures are able to generalize so well beyond the interpolation threshold), ML and Dynamical Systems (which will become increasingly important as artificial systems interact more and more with the physical world), and many more. If you want to see this project evolve, then I'd be delighted to have you along for the ride. I'm always open to suggestions of topics to cover.
Thank you for your time and happy learning!
I'm Native to the Iraqi Dialect but never learned Standard Arabic. Is there a point in spending effort learninf Standard Arabic or will people understand me regardless?
the use of non techie language, art style, and the cute side story of your character making their way though a dungeon, filled with puzzles and tasks ( building a PC ) Is an amazing way to engage first time builders!
Meta/Facebook AI introduces ‘Neural Prophet‘, a simple forecasting package that provides a solution to some of the most prevalent needs of customers, seeking to maximize the scalability and flexibility of time series forecasts based on Meta’s own internal data scientists and requests from external industry practitioners. Whether it’s estimating infection rates for disease management programs or projecting product demand to store inventory properly, the expanding data size necessitates new methodologies.
Machine learning methods that are nonparametric do not make any assumptions regarding the type of mapping function. They must be both accurate and simple to understand. By not forming hypotheses, they can choose any functional form from the training data. Thanks to this nature of deep learning models, they are scalable to match complex datasets. However, their black-box nature proves to be a disadvantage when projections are used to guide commercial or operational choices.
Paper: https://arxiv.org/abs/2111.15397?
Github: https://github.com/ourownstory/neural_prophet
https://preview.redd.it/0759uz4nxc381.png?width=1920&format=png&auto=webp&s=a7c623036bff8f47b9f2e712386c10b83bc87a91
Come on Google...Seriously?!?! If this is true, there is zero chance I buy anything Nest again. I've got a ton of these and a bunch of Nest Cams (which are also absent from the list). I was buying into their ecosystem, but definitely won't be doing that anymore if they don't handle this. If these were cheap, I'd get it...But they aren't. Throwing loyal customers to the curb is not cool:
https://www.tomsguide.com/news/your-nest-learning-thermostat-is-about-to-not-matter
Currently, I am using two mobile applications (for basic grammar) and a series of audio (for pronunciation and conversion of borrowed vocabularies) in studying Spanish. All of them are in English standard (English➡️Spanish). So far in my first week, I feel like I made so much progress.
Anyway, I just thought, because English is just our secondary language, it would be more easier if we use a learning material that directly uses Tagalog➡️Spanish (not Tagalog➡️English➡️Spanish). Since, English is not our native language, sometimes, I am having some lag issues during translation because my english vocabulary is not that extensive. I know that it would be more advantageous for us to use English➡️Spanish so we can also practice our fluency in English while learning Spanish. I am just wondering if you know any Tagalog➡️Spanish learning material, or if you are using one, is it indeed easier?
I recently had my first call with a good studio who was interested in hiring me for a project for a month. They had seen my work, liked it, but when the creative director described the project, requirements, etc, I realized two things:
-
There’s no way I could meet the project’s deadlines (I’m not yet at the professional skill level where I can work very quickly)
-
As I have no experience working for a studio/agency, I am clueless when it comes to professional workflows, how to work with a creative director, what types of questions to ask, what’s appropriate/professional and what’s not, what’s expected, etc. Basically I was totally in the dark. He was speaking to me as if I knew exactly what he was talking about (understandably) but in reality it was all completely new.
I ended up passing on the gig, with the intentions of upping my game, answering a lot of the questions I have, and learning to be able to work faster. I took a lot o away from the call, it was very valuable. But I’d like to know how to get the needed experience and learn more so I can do the job next time I have an opportunity like this.
Some context: I graduated university in 2020 and since then I’ve been working at a manufacturing company, making 3D product videos in c4D where I have no creative direction, I’m the only creative guy at the company, so no coworkers doing what I was doing, a boss who just wanted a cool final product and didn’t care how I made it, basically I was totally doing my own thing the way I wanted. It was really fun but I didn’t learn any common practices, good workflows, industry standards etc cause I was in a totally different industry (LED lighting fixtures).
So my question is how does one learn these things? Is it only by having studio experience that one learns how to work with creative directors, coworkers, producers, clients, etc? Is there a course that emulates some of those things? Are these things that are learned in an entry-level position at a studio, or internship? Can I learn just by doing smaller freelance gigs?
For those with lots of experience, how did you first learn these things? Any and all advice is deeply appreciated.
Thanks in advance
I've been learning a couple languages using this theory because it seemed to make a lot of sense to me and in my personal experience many of the claims it makes - such as around the effectiveness of memorisation vs simply hearing in context - seem to be true.
However I'm not entirely sure what the evidence is around the most effective way to learn a language or if there are any studies done at all.
Is Comprehensible Input Theory generally accepted as the best way to learn a language? Or at least as being better than the traditional way of learning? Or is it rather controversial? Maybe there are other learning methods that also work well?
What studies have been done around this and what were the results?
I basically just wanna know whether CIT really is as good as it seems to be (and whether the standard school system is as shit as I have been told)
Thanks so much!
I think Bitcoin is only half the picture for digital cash. The goal in my mind is to achieve the transaction use-case properly.
No matter the adoption rate of BTC scarcity will always compel it to be a SoV and volatility will prevent meaningful usage as a MoE. This is pretty much the flaw with the bitcoin standard argument in my view. It expects sats to be the UoA in the distant future. The issue is the cycle of hoarding will repeat itself. Doesn’t matter if everyone in the world knows or, loves and supports bitcoin. Their simply won’t be enough to go around, even with constant downward swings.
For the purposes of bringing cryptocurrency to the mainstream, this leaves fiat pegged stablecoins to fill in that void.
As such something possibly like an inflationary crypto currencywould be necessary in regards to its value against bitcoin for creating a hybrid coin so the underlying value could be appreciating at a greater rate than the percentage chosen. The inflation would then be a tool to disincentive hodling.
This would in return popularize Bitcoin more and increase its adoption and won't be a detriment.
Meta/Facebook AI introduces ‘Neural Prophet‘, a simple forecasting package that provides a solution to some of the most prevalent needs of customers, seeking to maximize the scalability and flexibility of time series forecasts based on Meta’s own internal data scientists and requests from external industry practitioners. Whether it’s estimating infection rates for disease management programs or projecting product demand to store inventory properly, the expanding data size necessitates new methodologies.
Machine learning methods that are nonparametric do not make any assumptions regarding the type of mapping function. They must be both accurate and simple to understand. By not forming hypotheses, they can choose any functional form from the training data. Thanks to this nature of deep learning models, they are scalable to match complex datasets. However, their black-box nature proves to be a disadvantage when projections are used to guide commercial or operational choices.
Paper: https://arxiv.org/abs/2111.15397?
Github: https://github.com/ourownstory/neural_prophet
https://preview.redd.it/013wyqjmxc381.png?width=1920&format=png&auto=webp&s=d50de48b1bb7bb9d85a92c3774cf15300a420cd6
Meta/Facebook AI introduces ‘Neural Prophet‘, a simple forecasting package that provides a solution to some of the most prevalent needs of customers, seeking to maximize the scalability and flexibility of time series forecasts based on Meta’s own internal data scientists and requests from external industry practitioners. Whether it’s estimating infection rates for disease management programs or projecting product demand to store inventory properly, the expanding data size necessitates new methodologies.
Machine learning methods that are nonparametric do not make any assumptions regarding the type of mapping function. They must be both accurate and simple to understand. By not forming hypotheses, they can choose any functional form from the training data. Thanks to this nature of deep learning models, they are scalable to match complex datasets. However, their black-box nature proves to be a disadvantage when projections are used to guide commercial or operational choices.
Paper: https://arxiv.org/abs/2111.15397?
Github: https://github.com/ourownstory/neural_prophet
https://preview.redd.it/qt7j80rlxc381.png?width=1920&format=png&auto=webp&s=82e781eecc48411c52657e1dbbd161dcce394cb0
