First things first — Enable Code Interpreter

To enable the code interpreter plugin, click on the three dots situated in the underside left corner and choose Settings.

A latest window will pop up. Select “Beta Features” and enable “Code Interpreter” as shown below.

After that open a latest chat and choose GPT-4. You must see the code interpreter listed within the dropdown.

Note: On July sixth, OpenAI announced that the code interpreter will probably be available to all ChatGPT Plus users over the subsequent week. In case, you continue to don’t have it yet, keep checking the Beta Features through the week.

Now let’s explore what we are able to do with the code interpreter.

ChatGPT’s Code Interpreter can generate a fast report and supply a link to download it in PDF format

Now that the code interpreter is enabled, we are able to upload a latest file. Within the chat bar, you need to see a latest plus icon. Click on it to upload this dataset in regards to the population of nations world wide from 1955 to 2020.

After uploading the file, you need to see this.

Here’s the prompt we’ll use to begin with a fast evaluation.

Analyze the dataset and make charts and graphs to indicate the key trends in population growth world wide

First, a fast descriptive statistic is finished where ChatGPT finds that there are 7 missing values and mechanically removes them from the dataset.

Then it continues with the info visualization.

And guess what? All of the code is generated inside ChatGPT. You simply must click on the “Show work” button to see the Python code generated.

But my favorite part is that you could export all this as a report in PDF format. Here’s the prompt that gets this done.

Answer it in the shape of a multiple pages PDF download

ChatGPT’s Code Interpreter can even make a customized report

Before we let the ChatGPT’s Code Interpreter select the visualizations for us, but now let’s do it ourselves and likewise add extra instructions to customize our report.

For this test, we’ll use a FIFA dataset with details about soccer players within the 12 months 2020.

Here’s the prompt we’ll use.

Act as a knowledge scientist and analyze the soccer players from only the countries United States, Canada, England, Brazil, and Argentina.

– Make a barplot to investigate the column “overall” which represents how good a player is at soccer.
– Make a histogram and a boxplot to explore the common height of players in such countries
– Make a scatterplot to see how the load of players is distributed.
– Finally, make a piechart to see the highest 10 Most worthy players (value_eur) from the USA.

First, as usual, it starts with a fast overview of the dataset. Something cool I notice is that it immediately selects the relevant columns for the evaluation.

Then it proceeds with the every visualization I asked for.