1
00:00:00,430 --> 00:00:03,013
（欢快的音乐）
(upbeat music)

2
00:00:05,160 --> 00:00:07,080
- 在这个视频中，我会给你
- In this video, I'm going to give you

3
00:00:07,080 --> 00:00:10,350
快速介绍我们的变压器模型
a very quick introduction to how our transformer models

4
00:00:10,350 --> 00:00:14,040
与 Tensorflow 和 Keras 一起工作。
work together with Tensorflow and Keras.

5
00:00:14,040 --> 00:00:15,510
非常简短的解释
The very short explanation

6
00:00:15,510 --> 00:00:17,310
是我们所有的 Tensorflow 模型
is that all of our Tensorflow models

7
00:00:17,310 --> 00:00:19,470
也是 Keras 模型对象，
are also Keras model objects,

8
00:00:19,470 --> 00:00:22,950
因此他们拥有标准的 Keras 模型 API。
and so they have the standard Keras model API.

9
00:00:22,950 --> 00:00:24,960
如果你是一位经验丰富的机器学习工程师
If you're an experienced machine learning engineer

10
00:00:24,960 --> 00:00:28,230
谁经常使用 Keras，这可能就是你需要知道的全部
who's used Keras a lot, that's probably all you need to know

11
00:00:28,230 --> 00:00:29,610
开始与他们合作。
to start working with them.

12
00:00:29,610 --> 00:00:30,900
但对其他人来说，
But for everyone else,

13
00:00:30,900 --> 00:00:34,170
包括那里的浪子 PyTorch 工程师
including the prodigal PyTorch engineers out there

14
00:00:34,170 --> 00:00:35,910
正在回归的人，
who are returning to the fold,

15
00:00:35,910 --> 00:00:38,430
我将快速介绍 Keras 模型，
I'm going to quickly introduce Keras models,

16
00:00:38,430 --> 00:00:40,440
以及我们如何与他们合作。
and how we work with them.

17
00:00:40,440 --> 00:00:43,080
在我将在下面链接的其他视频中，
In other videos, which I'll link below,

18
00:00:43,080 --> 00:00:46,440
我将更详细地介绍 Keras 模型的训练。
I'll run through training with Keras models in more detail.

19
00:00:46,440 --> 00:00:50,820
但首先，从高层次上讲，什么是 Keras 模型？
But first, at a high level, what is a Keras model?

20
00:00:50,820 --> 00:00:54,810
所以你的模型基本上包含了你的整个网络。
So your model basically contains your entire network.

21
00:00:54,810 --> 00:00:58,230
它包含层，以及这些层的权重，
It contains the layers, and the weights for those layers,

22
00:00:58,230 --> 00:01:00,690
并告诉模型如何处理它们
and also tells the model what to do with them

23
00:01:00,690 --> 00:01:02,880
所以它一路定义了整个路径
so it defines the whole path all the way

24
00:01:02,880 --> 00:01:05,460
从你的输入到你的输出。
from your inputs to your outputs.

25
00:01:05,460 --> 00:01:07,380
如果你以前使用过 Keras，
If you've used Keras before,

26
00:01:07,380 --> 00:01:09,480
你可能开始使用模型对象
you probably started using model objects

27
00:01:09,480 --> 00:01:11,850
通过手工构建它们，
by building them out by hand,

28
00:01:11,850 --> 00:01:14,250
你一层又一层地添加
you added one layer after another

29
00:01:14,250 --> 00:01:18,690
并且可能使用 model.add () 或功能方法。
and maybe using the model.add () or the functional approach.

30
00:01:18,690 --> 00:01:20,490
这并没有错。
And there's nothing wrong with that.

31
00:01:21,390 --> 00:01:23,430
许多伟大的模型都是以这种方式构建的
Lots of great models are built that way

32
00:01:23,430 --> 00:01:26,970
但你也可以预加载整个模型、权重和所有内容。
but you can also pre-load an entire model, weights and all.

33
00:01:26,970 --> 00:01:29,994
这真的很有帮助，因为如果你，
And this is really helpful, because if you,

34
00:01:29,994 --> 00:01:32,490
正如你在这里看到的，如果你尝试阅读这篇论文
as you can see here, if you try reading the paper

35
00:01:32,490 --> 00:01:34,110
或者如果你尝试查看代码，
or if you try looking at the code,

36
00:01:34,110 --> 00:01:37,350
你会看到 Transformer 的内部非常复杂，
you'll see the inside of a Transformer is pretty complex,

37
00:01:37,350 --> 00:01:40,110
从头开始把它写出来并把它做好
and writing it all out from scratch and getting it right

38
00:01:40,110 --> 00:01:41,850
即使对于有经验的人来说也很难
would be hard even for an experienced

39
00:01:41,850 --> 00:01:43,500
机器学习工程师。
machine learning engineer.

40
00:01:43,500 --> 00:01:45,870
但因为它都装在一个模型里，
But because it's all packed inside a model,

41
00:01:45,870 --> 00:01:48,150
你不需要担心那个的复杂性
you don't need to worry about that complexity on that

42
00:01:48,150 --> 00:01:49,140
如果你不想。
if you don't want to.

43
00:01:49,140 --> 00:01:51,570
如果你是一名研究人员，如果你想真正深入研究
If you're a researcher, if you want to really dig in there

44
00:01:51,570 --> 00:01:55,650
你可以，但你也可以只加载一个预训练的，
you can, but you can also just load a pre-trained,

45
00:01:55,650 --> 00:01:59,013
只需一行代码即可预配置变压器模型。
pre-configured transformer model in just one line of code.

46
00:02:00,150 --> 00:02:03,480
当我之前提到 Keras API 时，
And when I mentioned earlier about the Keras API,

47
00:02:03,480 --> 00:02:04,560
它的优点是
the advantage of it is that

48
00:02:04,560 --> 00:02:06,690
你是否从头开始编写自己的模型
whether you write your own model from scratch

49
00:02:06,690 --> 00:02:09,510
或者加载一个预训练的，你与模型交互
or load a pre-trained one, you interact with the model

50
00:02:09,510 --> 00:02:11,850
通过相同的 API，所以你使用
through that same API, so you use exactly

51
00:02:11,850 --> 00:02:13,950
同样的几种方法，你会看到它们
the same few methods and you're gonna see them

52
00:02:13,950 --> 00:02:16,380
一次又一次，这些方法就像适合，
again and again, these methods like fit,

53
00:02:16,380 --> 00:02:19,650
编译和预测，就像我提到的
compile and predict, and like I've mentioned

54
00:02:19,650 --> 00:02:22,530
我们将介绍如何使用这些方法的具体示例
we'll cover concrete examples of how to use those methods

55
00:02:22,530 --> 00:02:24,330
在我将在下面链接的视频中。
in the videos I'll link below.

56
00:02:24,330 --> 00:02:27,000
现在要从这个视频中拿走的关键是，
For now the key thing to take away from this video,

57
00:02:27,000 --> 00:02:28,950
如果你以前从未见过 Keras，
if you've never seen Keras before,

58
00:02:28,950 --> 00:02:30,870
是这种简洁的封装意味着
is that this neat encapsulation means

59
00:02:30,870 --> 00:02:33,090
一个巨大的神经网络的所有复杂性
that all the complexity of a huge neural net

60
00:02:33,090 --> 00:02:35,430
变得易于管理，因为你与它互动
becomes manageable, because you interact with it

61
00:02:35,430 --> 00:02:39,000
以完全相同的方式，使用完全相同的方法，
in exactly the same way, using exactly the same methods,

62
00:02:39,000 --> 00:02:41,700
是否是一个庞大的预训练语言模型
whether it's a huge pre-trained language model

63
00:02:41,700 --> 00:02:43,950
或你手写的简单模型。
or a simple model that you wrote out by hand.

64
00:02:45,466 --> 00:02:48,049
（欢快的音乐）
(upbeat music)