Pandas Min

Every dataset has a low end, and Pandas Min is the node that finds it for the whole table at once. Feed it a DataFrame, and it returns the minimum value of each column as a PDSERIES - one node, one output, "what's the floor here?" answered for every column in your data. It's the mirror image of Pandas Max, and the two of them together are the quickest way to sanity-check a freshly loaded table.

It's part of ComfyUI-Data-Analysis, Hide Inada's (HowToSD's) pack that turns pandas, Seaborn, and Matplotlib into ComfyUI nodes. The pack's whole idea is that structured data analysis should live in the same visual graph as image generation - and nodes like this one are the low-friction way to start exploring a dataset the moment it's loaded.

How it works

The node runs dataframe.min(). pandas returns the smallest value per column, skipping NaN by default. Column names become the Series index, so you can read the result at a glance and wire it anywhere. No options, no axis argument - column-wise min, always, exactly like dataframe.min() in a notebook.

Output is a PDSERIES. Usual routing applies: Pandas Series To DataFrame to make it a row, Pandas Series To String to display it, or straight into another Series-taking node.

The inputs that matter

Just dataframe. One input, one output, nothing to misconfigure.

How to install it

Same pack-wide install:

cd ComfyUI/custom_nodes
git clone https://github.com/HowToSD/ComfyUI-Data-Analysis.git
mv ComfyUI-Data-Analysis data-analysis
pip install -r requirements.txt

or ComfyUI Manager → search "Data analysis" → install → restart → reload. Dependencies are the pack's standard stack (pandas, matplotlib, seaborn, scipy, scikit-learn); no GPU, no models.

Common issues

• NaN for a column - the column is empty of numbers. All-missing or all-text columns return NaN instead of a value.
• A string "min" - pandas reports the alphabetically-first value of a text column as its min. If that reads wrong, the column needs a numeric conversion.
• Min and max look identical - you've got a constant column, or a rounding problem upstream. Worth a glance at the raw data.

It's the boring half of a useful pair. Load data → Pandas Min + Pandas Max → you instantly know the range of everything you're working with.