Pandas Show DataFrame

ComfyUI is a canvas built for images, so when you're running data through a pandas pipeline you can't see a thing unless you add a node that shows it. Pandas Show DataFrame is that node: it renders the DataFrame as text right on the node after every run and returns it as a STRING. Slap it on the end of any branch and you can finally verify your filtering, your cleaning, your pivot - instead of trusting the graph and hoping.

It's part of HowToSD/ComfyUI-Data-Analysis, and it's one of the pack's Display data nodes. The UI trick - text appearing directly on the node and being embedded into the workflow - is borrowed from pythongosssss's show-text nodes, and the pack credits that in its credit/ folder.

How it works

This is an output node, which means a few things mechanically. It calls str(dataframe) to produce the text, returns it as a list of STRINGs (that's the is_list output type), and - because it's an output node - it both stamps the text into the workflow's metadata and displays it in the node's UI. The is_output_node flag is what makes the difference from a plain converter: the result travels with the workflow and is visible on the canvas, which is exactly what "let me look at this" wants.

The output is pandas' standard string representation: index on the left, columns on top, truncated with ... in the middle for big tables. Long DataFrames show head and tail, not the whole thing - fine for verification, not a data export.

Inputs and output

• dataframe - the table to display.

Output: a list of STRING (the DataFrame as text).

Installation

Standard pack install. ComfyUI Manager: search Data analysis, install ComfyUI-Data-Analysis. Or:

cd ComfyUI/custom_nodes
git clone https://github.com/HowToSD/ComfyUI-Data-Analysis.git
mv ComfyUI-Data-Analysis data-analysis   # examples need this folder name
pip install -r requirements.txt

Restart, reload. Deps: pandas, matplotlib, seaborn, scipy, scikit-learn, openpyxl, lxml - no GPU, no models.

The gotchas

Big tables truncate, so don't mistake the preview for the whole dataset - if you need the full thing, slice it or save it with Pandas Save CSV. And because the output is text, the formatting is pandas' default: wide tables get wrapped and long values get ellipsized, which can be annoying to read but is honest. The genuinely useful pattern: drop one of these at every stage of a cleaning pipeline the first time you build it, then prune them once it works. Debugging data you can see is ten times faster than debugging data you're guessing about.