This holds true in most other countries as well. Gore/ chopping of appendages is happily accepted and enjoyed (in movies, games etc) by all of India, whilst a simple kiss can be a taboo/ issue.

Japan has some of the weirdest/inconsistent rules around this stuff. Black lines or mozaic partial censorship of genitals, incest/stuff with minors widely available, and then you have some pretty violent uncensored movies, manga/anime, and games (though while it's mostly a China thing, sometimes the blood gets censored to be white instead of red which doesn't actually make it better (also sometimes done for urine)), GTA5 is as popular there as anywhere, but game franchises like Mortal Kombat are banned.

And of course, even in America, we tend to like our violence and gore more over-the-top and simulated. Most people didn't care for liveleak type content, even fewer for not so hard to find footage from ongoing wars.

Japan runs custom scratch-built implementation of ethics reverse engineered from Western cultures. That's all. Consistence is key, but it's consistent only with itself, and nothing else, and explicitly not aligned to Christian religious scripts. Nothing Japanese is compatible with anything unless and until it is the sole dominant standard, like Sony storage media or Apple hardware. Always has been.

I made https://github.com/notAI-tech/fastDeploy for ease of use of deploying Deep learning models.

Except me and the team at my company nobody uses it.

After I made it lot of solutions came in this space, but IMO this is the easiest one to use still.

The model is trained by giving pairs of examples. So that it learns to generate an internal representation of images, texts or whatever inputs.

There are other ways to train, but this is the basic example.

2 chatgpt instances controlling the dual inputs would be funny

They both solve different issues. This is a library akin to huggingface transformers, while tensorflow js is akin to tensorflow or pytorch

Just wanted to drop in and appreciate your work.

It's documented very well, extremely easy to use and the pro version is priced very well.

> The 2021 M1 Air is a slightly chunkier fanless machine that runs circles around Intel laptops with desktop-class processors, at lower cost and up to triple/quadruple the battery life. It's unlike any Intel laptop you'll find on the market today.

Second this. I got the cheapest M1 macbook air in december and it really is streets ahead of competetion in terms of performace/ battery life.

The new macbook air is actually decent for gaming.

On the intel macbook airs (and on most pros without gpus) playing CS:GO at full resolution was a bad experience (overheating and <60fps).

I got the cheapest M1 macbook air and it plays CS:GO with no issues and without heating up (even when the ambient temperature in my city is ~38 Celsius)

any other games that you've played on them that you care to comment on performance wise? just curious

If you can position the camera so that only the "counter" is visible from it, a much easier option is to use the pre-trained yolov4.

https://gist.github.com/YashasSamaga/e2b19a6807a13046e399f4b... (download links for yolov4.weights is at https://github.com/AlexeyAB/darknet)

Using this, you will be able to detect if a cat is present in your image.

Length of given text after encoding it with their BPE. In general, you can expect it to be 1.2x-2x the len(words) in your text.