Name: 使用 Google Colab 進行數據科學和人工智能研究 (Using Google Colab for Data Science and AI)
Uploaded: 2025-04-18T07:34:59.000Z
Duration: 23 min 7 s

Are you tired of having to set up your Python coding environment every time you want to create a new coding or data science project?

您是否厭倦了每次要創建新的編碼或數據科學項目時都要設置 Python 編碼環境？

Well, what if there's a better way for you to do this right inside the browser and all that you need is an access to the internet?

那麼，如果有更好的方法讓您直接在瀏覽器中實現這一功能，而您所需要的僅僅是訪問互聯網，那又會怎樣呢？

Well, in this video, I'm going to show you how you could use Google Codelab to do that and this will unlock a lot of possibilities.

在本視頻中，我將向您展示如何使用 Google Codelab 來實現這一目標，這將為您帶來無限可能。

You'll be able to spin up a coding project or a data science project very easily with the click of a button.

您只需點擊一下按鈕，就能輕鬆啟動編碼項目或數據科學項目。

And yeah, let's have a look and let's dive in.

是的，讓我們來看看，讓我們來深入瞭解。

So the first thing that you want to do is head over to codelab.research.google.com and this will take you to the Welcome to Codelab page that you have here.

所以，你要做的第一件事就是前往 codelab.research.google.com，這會帶你進入這裡的 "歡迎使用 Codelab "頁面。

You'll see here on the Go that it's providing you with instruction on how to access Google's LLM model called Gemini.

你會在 Go 上看到，它正在指導你如何訪問名為 "雙子座 "的谷歌 LLM 模型。

It tells you the step-by-step in order to do so as well.

它還會告訴你如何一步一步地進行操作。

So here it summarizes Google Codelab as kind of like a coding environment that allows you to write and execute Python right inside the browser.

是以，它將 Google Codelab 概括為一種類似於編碼環境的工具，可以讓你在瀏覽器中編寫和執行 Python。

So if you are already familiar with Jupyter Notebooks, same thing, but then on Google Cloud and it's hosted there.

是以，如果你已經熟悉了 Jupyter Notebooks，那也是一樣的，但它是在谷歌雲上託管的。

And great thing is that you'll be able to access this for free.

Do note that there are also paid options for you if you need more compute power or compute time.

請注意，如果您需要更強的計算能力或計算時間，也可以選擇付費選項。

A great thing about the Google Codelab is that you could get up and running fairly quickly in a few seconds and you'll be also be able to share the resulting notebook to your peers and colleagues.

谷歌 Codelab 的一大優點是，您可以在幾秒鐘內快速啟動和運行，還可以將生成的筆記本與同行和同事共享。

So here you'll immediately notice that there's option to include text in your notebook and also the coding aspect in your notebook as well.

在這裡，你會立即發現有一個選項可以在筆記本中加入文本，還可以在筆記本中加入編碼方面的內容。

And you'll notice that this text is stylized by Markdown, which is a syntax that allows you to do, you know, enter text in here, allows you to stylize it as a bold or italic, or also allow you to add links to your notebook.

你會注意到，這些文本是用 Markdown 來進行樣式化的，這是一種允許你在這裡輸入文本的文法，允許你將文本樣式化為粗體或斜體，也允許你在筆記本上添加鏈接。

And so the here, you know, it's showing an example of creating a Matplotlib chart.

是以，這裡展示了一個創建 Matplotlib 圖表的示例。

And so why don't I show you how I would use Google Codelab?

那麼，我為什麼不向你們展示一下我是如何使用 Google Codelab 的？

So to spin up a new notebook, you'll click on the file tab menu here, click on new notebook in drive, and it'll spin up a new notebook for you.

要啟動一個新筆記本，點擊這裡的文件標籤菜單，點擊驅動器中的新筆記本，它就會為你啟動一個新筆記本。

So under the hood is actually a compute that is spinning up for your use case.

是以，在引擎蓋下，實際上是一個正在為你的用例旋轉的計算器。

So here we're to a remote computer and for the name of the file is actually being stored on your Google drive.

是以，在這裡我們連接的是遠程計算機，文件名實際上存儲在谷歌硬盤中。

So great thing is after your coding session, you'll always have access to your code.

最棒的是，在編碼會話結束後，您可以隨時訪問您的代碼。

So even if you close your browser, and if the notebook has saved your code, it should be safe in your Google drive.

是以，即使您關閉了瀏覽器，如果筆記本保存了您的代碼，它也會安全地保存在您的 Google 驅動器中。

And then you could access that on another computer, which might be handy when you're on the go, if you're using your phone, or if you're using your tablet, or if you're using your work laptop, right.

然後，你就可以在另一臺電腦上訪問它，這在你外出、使用手機、平板電腦或工作用筆記本電腦時可能會很方便，對吧。

And then when you go home, you have your desktop.

回家後，你就有了自己的桌面。

Or if you go to the office, you have your desktop, which will allow you to access the same version if you're using the same Google account.

或者，如果你去辦公室，你有你的臺式機，如果你使用相同的谷歌賬戶，就可以訪問相同的版本。

我就稱它為我的第一本筆記本吧。

And here you have the option to create a code cell or a text cell.

在這裡，你可以選擇創建代碼單元格或文本單元格。

So by default, it gives you a coding cell here.

是以，默認情況下，這裡會給你一個編碼單元格。

And here, you're going to see the text cell, it gives you the option to stylize your markdown text.

在這裡，你會看到文本單元格，它為你提供了對標記符文本進行風格化的選項。

So here, if we click once, you'll be able to see the heading with one hash symbol, which signifies heading one.

是以，在這裡，如果我們點擊一次，你就能看到標題上有一個哈希符號，表示標題一。

So if I want to run this, I'll just, you know, hit on shift, enter, and then this will run.

是以，如果我想運行它，只需按下 shift 鍵和回車鍵，它就會運行。

And then you'll see that it is not in the editing mode, it is now in the view mode.

然後你就會看到，它現在不是在編輯模式下，而是在視圖模式下。

If I click on another text cell, and then let's have two heading here, let's run.

如果我點擊另一個文本格式，然後在這裡設置兩個標題，讓我們運行一下。

Same as hitting shift and enter at the same time.

就像同時按下 shift 和 enter 鍵一樣。

And if you click on the table of contents, you'll be able to see the new section here.

如果你點擊目錄，就能看到這裡的新章節。

How about I'll call it my first notebook.

我把它叫做我的第一本筆記本怎麼樣？

And then you'll see that my first notebook here is the first item on the table of contents.

然後你會看到，我的第一本筆記本是目錄上的第一個項目。

You'll notice that the new section here is a sub of the my first notebook.

你會發現，這裡的新部分是我第一本筆記本的子部分。

So why don't I create a new one called um, example, example.

所以，我為什麼不創建一個新的，叫做嗯，例子，例子。

All right, and new section here will be h2 with two as two hash symbol, sorry, two hash symbol here will signify that it is a h2 heading.

好的，這裡的新部分將是 h2，帶有兩個哈希符號，抱歉，這裡的兩個哈希符號將表示這是一個 h2 標題。

And if you have one hash symbol, it is a h1 heading.

如果有一個哈希符號，它就是 h1 標題。

It means that heading at level one, subheading at level two, right?

也就是說，一級標題，二級副標題，對嗎？

So new section here will belong under example section.

是以，這裡的新部分將屬於示例部分。

If you add another to be with triple hash symbols, then this will be h3.

如果您再添加一個三重哈希符號，那麼這將是 h3。

So you have one at one hash, two hash, three hash, and then it'll be displayed hierarchically like that.

這樣，你就有了一個哈希值、兩個哈希值、三個哈希值，然後它就會像這樣分級顯示。

Let me create another section, call it examples.

讓我再創建一個部分，稱之為 "示例"。

And so in our first example, let's have a look at using pandas.

是以，在第一個例子中，讓我們來看看如何使用 pandas。

So by default, pandas is already pre installed in your notebook.

是以，默認情況下，你的筆記本中已經預裝了 pandas。

So in order to verify that let's do import pandas as PD, and then we'll hit enter.

為了驗證這一點，讓我們以 PD 的形式導入 pandas，然後點擊回車鍵。

All right, so it is able to run without an issue.

Let's have a look at what if you want to install other Python libraries that might not be in the notebook?

我們來看看，如果要安裝筆記本中可能沒有的其他 Python 庫，該怎麼辦？

好吧，為了做到這一點，讓我來看看這個。

I think it's ytdlp, which is a video downloader.

我認為是 ytdlp，它是一個視頻下載器。

So let's go to the official GitHub repo as well.

所以，讓我們也去 GitHub 官方倉庫看看吧。

But before that, let's make note of this.

但在此之前，我們先來了解一下。

是以，這裡實際上有很多代碼。

So in order to make it quicker, let me go to a prior notebook that I've created using assembly AI, the example.

是以，為了更快捷，讓我以之前使用 AI 程序集創建的筆記本為例。

So in this video, we pretty much downloaded a YouTube video.

是以，在這段視頻中，我們幾乎下載了一個 YouTube 視頻。

所以 ytdlp 就是這個名字，對嗎？

So you'll notice here that to use the library, it is called yt underscore DLP.

所以你會注意到，要使用這個庫，它的名稱是 yt underscore DLP。

But then to install it, it is yt dash DLP.

但要安裝它，則需要 yt dash DLP。

Okay, so note that it's slightly different in spelling.

So let's skip Panda's example here for a moment.

是以，我們暫且跳過熊貓的例子。

Notice that after we hit on the shift enter for the text cell, a new coding cell is created automatically, which is quite helpful because we want to type in code now.

請注意，當我們按下文本單元格的 shift enter 鍵後，會自動創建一個新的編碼單元格，這非常有用，因為我們現在要輸入代碼。

是以，我們要使用 pip 安裝 yt。

Oh, and it pretty much detects that we want to do so as well.

哦，它還能檢測到我們想這樣做。

So let's run it, you know, either shift enter or hit on the play button here.

所以，讓我們運行它，你知道，要麼換擋回車，要麼點擊這裡的播放按鈕。

And so yeah, you'll notice that when you're running this, we have this status here on the RAM and the disk hard drive consumption.

是的，你會注意到，當你運行這個程序時，我們會顯示內存和硬盤消耗的狀態。

So you'll notice that we have access to a Python three environment with almost 13 gigabytes of RAM and more than 100 gigabytes of storage for running your notebook.

是以，你會發現我們可以訪問 Python 3 環境，該環境擁有近 13GB 內存和 100GB 以上的存儲空間，可用於運行你的筆記本電腦。

So you could even run a pretty powerful workflow using, you know, your mobile device as well.

是以，你甚至可以使用移動設備運行一個相當強大的工作流程。

Now that we have ytdlp installed, let's have a look at some of the code here.

現在我們已經安裝了 ytdlp，讓我們來看看這裡的一些代碼。

We'll copy this code block and then paste it in here.

我們複製這個代碼塊，然後粘貼到這裡。

And let's copy the now here, paste it, then run the function here.

現在，讓我們在這裡複製、粘貼，然後在這裡運行函數。

And it's suggesting us to do this following.

I'll just hit on tab, and then I'll run it shift enter or just click on the play button.

我只需點擊選項卡，然後運行移位回車或點擊播放按鈕即可。

You'll see that the file is being downloaded from YouTube, which is the video on Steve Jobs 2005 Stanford commencement address.

你會看到文件是從 YouTube 下載的，這是史蒂夫-喬布斯 2005 年斯坦福大學畢業典禮演講的視頻。

In the meantime, we'll go here to the files and it's downloading.

與此同時，我們將在這裡下載文件。

字幕列表影片播放

使用 Google Colab 進行數據科學和人工智能研究 (Using Google Colab for Data Science and AI)

essentially

access

familiar

bold