Getting Started with Google Colab

This guide provides instructions to create a a Google Colab file and use its AI tools to managed sample datasets.

Tasha Penwell
68 steps
8 minutes

Google

While business majors aren't expected to become software developers, familiarity with Google Colab and its AI tools offers a significant advantage in today's data-driven world. Here's why: - **Data Analysis & Visualization:** Colab provides a user-friendly environment to analyze and visualize data, crucial skills for understanding market trends, consumer behavior, and making informed business decisions. - **AI-Powered Insights:** Colab integrates with powerful AI tools that can automate tasks, generate reports, and provide predictive analytics, giving business professionals a competitive edge in strategic planning. - **No-Code Solutions:** Colab offers tools that require minimal coding knowledge, allowing business majors to leverage AI capabilities without needing to be programmers. - **Collaboration & Innovation:** Colab facilitates collaboration and allows for the exploration of new ideas and solutions using AI, fostering innovation in business practices. By understanding the potential of Google Colab and its AI tools, business majors can enhance their analytical skills, gain valuable insights, and drive innovation in their future roles, even without being developers themselves.

Remember, the goal is not be a carpenter but know enough to identify a phillips from a flathead screwdriver.

Google Colab lets you write and run computer programs (specifically in Python, which is popular for data analysis) without needing to install anything on your own computer. Think of it as a shared online document where you can type instructions and see the computer's responses. It's especially useful for working with data and doing things like analyzing trends or making predictions, which can be helpful in business.

Setting Up the Plugins

For this demo, I am using a Google account that has minimum added features in an unpaid account. You should have access or be able to do everything you see in the following steps.

Plugins are like add-ons for software. They give the software extra features. The "Colaboratory" and "Sheets to Colab" plugins help you easily move files between Google Drive (where you might store data in Sheets) and Google Colab (where you'll analyze that data with code). They make it much simpler to get your information into Colab so you can work with it.

Go to Google Drive and select **+New**

Click **More > + Connect more apps**

The Google Workspace Marketplace opens.

Do a search for *Colab*

Find and select the *Colaboratory* tool

Click **Install**

Click **Continue**

You can uncheck this box if you prefer not to have this as your default app. It is up to you. Click **OK**

Click **Done**

You are now going to search for another plugin. In the search field type *Sheets to Colab*

Select **Sheets to Colab**

Click **Install**

Click **Continue**

Click **Done**

Close out he Google Workspace Marketplace

Create a Google Colab File

In your Google Drive, click **+New**

Click **Google Colaboratory**

You are now in Google Colab!!! Just like other Google resources, this is a cloud based application that requires access to the Internet and auto saves your work Close out the Release notes on the right side,

Creating Sample Data Sets using Gemini

Open a new tab

Search for the Google Gemini AI tool

In the prompt, type "CREATE 2 TABLES THAT ARE RELATED TO MUSIC INDUSTRY WITH AT LEAST 10 RECORDS. THESE WILL BE USED FOR AN INNER JOIN EXERCISE, SO THE FIELDS NEEDS TO BE APPROPRIATE FOR AT LEAST ONE ATTRIBUTE TO JOIN ON " and view the results.

Individual results will vary but you should something similar to the **Table 1 Artists** and **Table 2 Album** as shown in the next two screenshots.

Click on the **Export to Sheets** link for both tables.

You can use the pop up to **Open Sheets** or go straight to your Google Drive and find the spreadsheet files created.

Viewing and Renaming the Spreadsheets

When you create sample datasets and export them to Google Sheets, you'll notice the sheet and tab names are based on the prompt used to generate the data. This is just a default naming convention. You'll probably want to rename the sheet tabs and the overall file to something more descriptive and relevant to your project. It's a quick step, but it will make your data much easier to manage and understand later on.

Update the file name to something more appropriate. In this demo, Column A tells me this is the Album table so...

I am renaming to *Albums.* Rename the sheet tab to appropriate name as well. I renamed mine to *Albums*

For the second file, I am going to go to **File > Open**

Find the file - again the name of the file is the prompt used.

Click **Open**

Right-click here

Rename the file and tab as you did the other file - use an appropriate name. In this demo, Column A tells me this is about Artist so I am naming the file and tab *Artist.*

Keep these spreadsheets open. You will need them in this exercise.

Return back to the Google Colab tab

Rename the Colab file to something more appropriate based on the tables. In this demo, I am renaming it to Album_Artist. This is done just as you do in other Google files (click on it and rename it). Keep the extension of [[.ipynb]]

Add the Data to Your Colab Notebook

Copy the URL in the address bar from one of your spreadsheets. In this demo, my first spreadsheet is the *Albums* spreadsheet.

Go back to the Colab file.

Click the **generate** link to begin using the AI tool.

Accept the terms by clicking **Continue**

You see that there is a Generate box waiting for your prompt. Enter the below prompt ***import the data from* [[\[insert the URL from your spreadsheet\\]]\] *make it a table that will be used as a join. name the dataframe to identify if as data on Albums*** Hit [[Enter]] We see that the AI tool writes the code for you based on your prompt. Spend a few minutes looking at it. Notice that the [[spreadsheet_url]] is your spreadsheet file.

A dataframe is like a digital spreadsheet or table. It organizes data into rows (like individual entries) and columns (like categories or features). Think of it as a way to neatly present and work with data, especially when you have a lot of it.

The dataframe is named *[[df_albums]]* which helps us identify what this dataframe consists of. The AI tool did this thanks to our explicit prompt

The code is written - now we need to see if it works as expected! (fingers crossed) To run the code, click on the button that looks like a play icon next to the code. This will run the code that is in this specific block.

Follow the prompts to allow access to your Google credentials. This includes...

Signing in...

Continuing...

Checking the box...

Continuing...

Once you have agreed and allowed everything, you should be back in your Colab file and see a table from the data in your spreadsheet. Something go wrong? Do not fret - just ask me for help!

Adding Data from the Second Table

We are going to use AI to help bring our second table into this Colab notebook. Click on the **generate** link

In the prompt type "*create the second table for Artists. The sheet id is* " and then go to your second spreadsheet file

You are going to select the sheet id to copy and paste. The sheet id can be found in URL, but it is only a portion of it. You want to copy the portion after the *[[d/ ]]all the way to right before the [[/edit.]]* In this demo, you can see mine is [[1Ez8tqspElxJQtZWTGP3ownFIU0n61BqYaAybeANlBeQ]]. Yours will be different - so do not copy mine please.

Go back to your Colab file and finish the prompt by pasting your sheet ID number. Finish the prompt by asking AI to name this new dataframe so it represents data having Artists.

AI writes the code based on your prompt. Spend a few minutes inspecting the code. Notice that it has data representing this second table (Artists in this demo). Now, we need to run the code to see if it works as expected.

Hopefully, it produces a table similar to this! If not, ask me for help!

Joining the Two Tables Together with an Inner Join

An inner join is a type of join operation in SQL that combines rows from two or more tables based on a common column or columns. The join condition specifies how the rows from the tables are matched. Only rows that satisfy the join condition are included in the result set. Inner joins are useful when you want to retrieve data from multiple tables that are related to each other. In this example, we have a table of Artists and a table of Albums. The INNER JOIN ensures that only albums that have a corresponding artist in the *Artists* table and vice-versa are included in the result. Basically, it is used to selects records that have matching values in both tables. Joins can be helpful when working with individual datasets as you are in your projects. If you see some similar attributes (columns) in two different tables and want to learn how to create a new table from these two different tables, joins can be helpful. We do not cover this in class because it is a little higher level than what we cover BUT you can ask me! Also, you can learn more about joins at <https://www.w3schools.com/sql/sql_join_inner.asp>

Click the **+Code** button to add a new code box.

Click the **generate** link to use the AI feature

Type the prompt "create an inner join based on the ArtistID field from these two tables". The code is generated based on the prompt. The *ArtistID* attribute was the common attribute (column) that both tables had. This is the attribute we are basing our inner join on. If you did something different, find the column name that is the same (casing, spacing, etc.) in both tables. Run the code by clicking the play button in this code block.

We now have a new table combining fields from both individual tables!

Additional Features!

Next to this new table, we see a couple of additional features we can use. Explore these tools to lean more about what Colab offers.

Build the Bridge

How can combining datasets help find trends in your research? Submit one document answering this questions into your Teams channel with a statement (a sentence or two is fine) from everyone on the team.

Getting Started with Google Colab

This guide provides instructions to create a a Google Colab file and use its AI tools to managed sample datasets.

Tasha Penwell
68 steps
8 minutes

Google

Remember, the goal is not be a carpenter but know enough to identify a phillips from a flathead screwdriver.

Setting Up the Plugins

For this demo, I am using a Google account that has minimum added features in an unpaid account. You should have access or be able to do everything you see in the following steps.

Go to Google Drive and select **+New**

Click **More > + Connect more apps**

The Google Workspace Marketplace opens.

Do a search for *Colab*

Find and select the *Colaboratory* tool

Click **Install**

Click **Continue**

You can uncheck this box if you prefer not to have this as your default app. It is up to you. Click **OK**

Click **Done**

You are now going to search for another plugin. In the search field type *Sheets to Colab*

Select **Sheets to Colab**

Click **Install**

Click **Continue**

Click **Done**

Close out he Google Workspace Marketplace

Create a Google Colab File

In your Google Drive, click **+New**

Click **Google Colaboratory**

Creating Sample Data Sets using Gemini

Open a new tab

Search for the Google Gemini AI tool

Individual results will vary but you should something similar to the **Table 1 Artists** and **Table 2 Album** as shown in the next two screenshots.

Click on the **Export to Sheets** link for both tables.

You can use the pop up to **Open Sheets** or go straight to your Google Drive and find the spreadsheet files created.

Viewing and Renaming the Spreadsheets

Update the file name to something more appropriate. In this demo, Column A tells me this is the Album table so...

I am renaming to *Albums.* Rename the sheet tab to appropriate name as well. I renamed mine to *Albums*

For the second file, I am going to go to **File > Open**

Find the file - again the name of the file is the prompt used.

Click **Open**

Right-click here

Rename the file and tab as you did the other file - use an appropriate name. In this demo, Column A tells me this is about Artist so I am naming the file and tab *Artist.*

Keep these spreadsheets open. You will need them in this exercise.

Return back to the Google Colab tab

Add the Data to Your Colab Notebook

Copy the URL in the address bar from one of your spreadsheets. In this demo, my first spreadsheet is the *Albums* spreadsheet.

Go back to the Colab file.

Click the **generate** link to begin using the AI tool.

Accept the terms by clicking **Continue**

The dataframe is named *[[df_albums]]* which helps us identify what this dataframe consists of. The AI tool did this thanks to our explicit prompt

Follow the prompts to allow access to your Google credentials. This includes...

Signing in...

Continuing...

Checking the box...

Continuing...

Once you have agreed and allowed everything, you should be back in your Colab file and see a table from the data in your spreadsheet. Something go wrong? Do not fret - just ask me for help!

Adding Data from the Second Table

We are going to use AI to help bring our second table into this Colab notebook. Click on the **generate** link

In the prompt type "*create the second table for Artists. The sheet id is* " and then go to your second spreadsheet file

Go back to your Colab file and finish the prompt by pasting your sheet ID number. Finish the prompt by asking AI to name this new dataframe so it represents data having Artists.

Hopefully, it produces a table similar to this! If not, ask me for help!

Joining the Two Tables Together with an Inner Join

Click the **+Code** button to add a new code box.

Click the **generate** link to use the AI feature

We now have a new table combining fields from both individual tables!

Additional Features!

Next to this new table, we see a couple of additional features we can use. Explore these tools to lean more about what Colab offers.

Build the Bridge