Do you know if he'll be doing the full color octet calculations?

你知道他会不会教矩阵乘法做八色胶子运算吗？

The most important kind of magical step is laying them out in a matrix and doing what's called matrix multiplication.

最关键的神奇步骤是将它们排列成矩阵，并进行所谓的矩阵乘法。

So, if you wanna build any AI application, you have to have an accelerator, just like we have in the cloud, to be able to run these matrix math multiplications fast in order to drive AI features.

如果你想构建任何人工应的话，你必须拥有一个像云端那样的加速器，便够快速运行这些矩阵数学乘法，从而驱动人工。

So you can get away with a smaller matrix multiply unit.

所你可一个较小的矩阵乘法单元。

And indeed, this is how you would tell a computer to perform quaternion multiplication, and the relative compactness of this operation compared to, say, matrix multiplication, is what's made quaternions so useful for graphics programming and many other things.

确实，这就是你告诉计算机执行四元数乘法的方式，而与矩阵乘法相比，这种操作的相对紧凑性得四元数在图形编程及其他许多领域中非常有。

One place where he didn't enough was we missed a matrix multiply operator, like that finally got added to Python, but about 10 years later than it should have.

有一个地方他做得不够， 就是我们错过了矩阵乘法运算符，这个最终被添加到Python中，但比应该的时间晚了大约10年。

I can make matrix multiply go 10 times faster.

我可将矩阵乘法运算速度提升十倍。

Once you know that it's linear, you know that there's some way to describe this function as matrix multiplication.

一旦你知道它是线性的 你就知道有一些方法可矩阵乘法来描述这个函数。

And the reason they made the vocabulary so large is because TPUs like matrix multiply unit is massive.

他们将词汇量设计得如此庞大，原因在于像TPU这样的矩阵乘法单元规模巨大。

It boils down essentially to a pair of matrix multiplications with a simple something in between.

归根结底就是两次矩阵乘法，中间夹了一个简单的操作。

All these chips have big matrix multiply units.

所有这些芯片都有大型矩阵乘法单元。

And then 3.5 as the matrix multiply operator, and now it's like, okay, we gotta use that.

然后3.5作为矩阵乘法运算符，现在我们得它了。

But when you think about matrix multiplication as applying one transformation after another, this property is just trivial.

但是当你把矩阵乘法看作是应一个又一个变换时 这个性质就很简单了。

Unfortunately, matrix multiply is not the way in which we tackle problems today.

遗憾的是，矩阵乘法并非我们如今解决问题的方式。

This is an honest goodness proof that matrix multiplication is associated, and even better than that, it's a good explanation for why that property should be true.

这是一个很好的证明矩阵乘法是相关的 甚至比这更好 它很好地解释了为什么这个性质是正确的。

So let's start by rewriting that right hand side as some kind of matrix vector multiplication, using a matrix which has the effect of scaling any vector by a factor of lambda.

我们先把右边写成某种矩阵向量的乘法 一个矩阵它的作是将任意向量乘的倍数。

So matrix multiply has the ability of moving n squared pieces of data and doing n cubed operations on it.

因此，矩阵乘法够移动n平方个数据片段，并对这些数据进行n立方的运算。

So the benchmark has at its core, a matrix multiply, two matrices being multiplied together.

因此，基准测试的核心是矩阵乘法，即两个矩阵相乘。

You could even define this as matrix vector multiplication when you put the matrix on the left of the vector, like its a function.

你甚至可把它定义为矩阵向量乘法当你把矩阵放在向量的左边 就像它是一个函数。

Trust me, this will give you a much better conceptual framework that makes the properties of matrix multiplication much easier to understand.

相信我 这将给你一个更好的概念框架 矩阵乘法的性质更容易理解。