Enhance the data viewer for Pandas dataframes in VSCode

[rchiodo]

Discussed in #9867

^{Originally posted by xareelee April 28, 2022}
When you run Jupyter/IPython in VSCode, you can view the Pandas dataframes variable in the data viewer:

Feature requests for data viewer:

1. add info as info() in the data viewer

• dtype for each column
• non-null count
• unique count
• total memory usage

as df.info():

2. add a button to view describe() for columns

• you can select columns to show stat description.
• when pressing the button, display a popup view to show the results of describe() for selected columns (or all columns if there is no selection)

as df.describe():

3. provide a textfield to input the expression to apply onto the dataframe

You can add fluent dot syntax onto the dataframe and view the result on the fly. It's useful for data tidying and cleaning.

The textfield looks like:

# The below line is the input in the textfield; `(df)` is a prompt for the variable name
(df) {this is a placeholder for hints}

For example, you can type .max() in the textfield, it will update the data viewer for the output:

(df) .max()

You can chain the methods the with fluent dot syntax:

(df) .astype({'Open': 'int32', 'High': 'int32', 'Low': 'int32', 'Close': 'int32'}).drop(columns=['Volume'])

Those would be very useful to view the Pandas dataframes in VSCode.

[jmew]

This feature might be covered by Data Wrangler. Every feature that was requested here can be done in Data Wrangler today already.

[rchiodo]

I thought it might :)

Do you have any public information you can share about an ETA?

[jmew]

@rchiodo We are hoping to get an VS Code Insiders release in June but cannot share the repo at this time as it will be private until the insiders release. I think we can use this issue or the original discussion issue if users have any other comments or feedback for what else they'd like to see in this "enhanced data viewer".

[Philip-S-Martin]

@jmew is there any way I can track the progress on the data wrangler?

[nickdos]

This feature might be covered by Data Wrangler. Every feature that was requested here can be done in Data Wrangler today already.

I'm trying to solve the same issue and I can't find "Data Wrangler" via Google (just seems to give me the AWS version of "data wrangler" or general blogs on Python "data wrangling"). Can you provide a link to this, please?

[EHardiR]

I know multiple teams (myself included) that refuse to start using VS Code until there is a better way to view and filter dataframes in Pandas. For now, they stick to PyCharm and RStudio.

[DonJayamanne]

Duplicate of #1148