For every drink you down, you're likely to urinate 100 mL more than you consumed.

你每喝一口酒，排尿量就可能比摄入量多100毫升。

Now, as the frames come in from the Camera preview, we simply create a Core ML request to run on each of these frames.

相机预览的各帧出现时，我要建立一个 Core ML 请求，让它在每一帧上运行就可以了。

Now at any given moment, there's about 150 ml of CSF in the body.

在任何特定时刻，体内有大约150毫升的脑脊液。

So if you have two tubs of product, both 500 ml, the ice cream is gonna weigh about 300 grams.

因此如果你有两个500毫升的产品桶，冰淇淋大约重300克。

The total volume may be closer to 110 ml, and 70ml is the fraction that got ejected out with each beat, the other 40ml kind of lingers in the left ventricle until the next beat, right?

总容积量可能更接近于110毫升，而70毫升是每次心脏搏动喷出的血量，那么剩余的40毫升留在左心室直到下一次心脏搏动，吧？

In a study of elderly adults with memory problems, 500 ml of purple grape juice per day for twelve weeks allowed them to learn more words compared to a placebo group.

在一项针有记忆问题的老年人的研究中，与安慰剂组相比，连续 12 周每天喝 500 毫升紫葡萄汁能让他学会更多的单词。

A full bladder can comfortably hold around 500 ml of pee, though it can usually expand to hold a maximum of around one liter.

一个饱满的膀胱可以舒适地容纳大约 500 毫升的尿，尽管它通常可以膨胀到最多容纳大约一升。

Okay, action figures. Nice. ML Legends Avenger iron spider Spidey Tom Holland six, loose action figure.

下一个，手办。不错！ML传奇复仇者铁侠蜘蛛侠 汤姆·赫兰德，六寸。已开箱手办。

For most people, drinking a small amount of milk, let's say 150 ml will not cause issues.

于大多数人来说，喝少量的牛奶不会造成问题。

And most importantly, MPS also enables everyone using the powerful, new Core ML framework that Seb just described to get the full performance of the GPU for machine learning automatically.

最重要的是，MPS 也让每个人能够使用小塞刚才提到的强力的新的 CoreML 框架，因而获得 GPU 的全部性能以供自动机器学习。

I like it also because typically a cup, an average cup is about 250 mL of brewing or about eight ounces and that scales really nicely.

我喜欢它的另一个原因是，一般一杯的冲泡量大约是250毫升或8盎司，这非常合适。

This is constantly replenished, with around 500 ml of new CSF produced everyday and the excess, or 500 minus 150 mL or 350 mL, is absorbed into the blood.

这是不断补充的，每天我能产生大约500毫升的新脊髓液；多余的脊髓液，也就是500减去150毫升等于350毫升的脊髓液，会被吸收进入血液。

And one of those is actually the integration between Vision and Core ML.

其中一个就是 Vision 和 Core ML 的整合。

In Italy, a cappuccino will often be in a relatively small cup and contain only around 100 ml of milk.

在意大利，一杯卡布奇诺通常装在一个相较小的杯子里，含有大约 100 毫升的牛奶。

I can't do the maths on that, if I need 500 mL of coffee, I'm not dividing 500 by 15 first thing in the morning.

我不能算出来，如果我要500毫升咖啡，我不会在早上的第一件事就是用500除以15。

Core ML itself provides a rich set of primitives to integrate state-of-the-art machine learning models into your apps.

Core ML 本身就提供丰富的原语，来把当前最新的机器学习模型整合到你的应用程序中。

You then take that model and run it through the Converter tool, and that Converter tool will produce a Core ML compatible format.

然后你在转换器工具中运行那个模型，那个转换器工具产生一个可在 Core ML 兼容的格式。

So when you bring all of this together, Core ML really allows you to easily leverage the power of machine learning to create truly intelligent applications.

把所有这些加在一起，Core ML 真的能让你轻松地利用机器学习的力量制作出真正智能的应用。

Now, we're also open sourcing the Core ML Converter tool so that we can get even broader community adoption over time.

我还开放了 Core ML 转换器工具的源代码，这样在未来它就能更广泛地被使用了。

In a 12 oz (355 ml) can of Coke there are 39 grams of sugar.

355 毫升） 的可乐中含有 39 克糖。