Improve the display of visualize

[MarieSacksick]

When using the following piece of code in vscode in a python file with the extension to be able to execute cell per cell:

# %%
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split

X, y = load_iris(return_X_y=True, as_frame=True)
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state=42)

# %%
from sklearn.tree import DecisionTreeClassifier

model = DecisionTreeClassifier(random_state=42)
model.fit(X_train, y_train)

# %%
from skops import io as sio
sio.dump(model, "model.skops")

# %%
sio.visualize("model.skops")

The output is not really readible, there is plenty of spaces between each line, and the indentation is not correct (cf screen shot below).

It affect the usability and interest of this feature. If the display could be improved, it would be awesome!

[BenjaminBossan]

This seems to be an issue with the visualization in jupyter notebooks, in the terminal the same code looks nicely formatted:

[adrinjalali]

It seems it'd be nice to figure out why it breaks in notebooks and try to fix it, but I'm not sure how.

[divakaivan]

Could you recreate this behaviour? Might be some env/setup thing that's different 🤔

I tried the code in:

• google colab notebook - works fine ✅
• notebook opened in vscode - works fine ✅
• notebook opened through jupyter - works fine ✅
• python file with the cell run extension (what Marie shared)- works fine ✅

Evidence

• google colab notebook

• notebook opened in vscode

• notebook opened through jupyter

• python file with the cell run extension

[MarieSacksick]

thanks @divakaivan, your tests pushed me to try again with a new conda env and all, and it displays properly.
BUT, when I use rich, it starts failing.

Steps:

• conda create -n skops python=3.13
• conda activate skops
• pip install pandas, ipykernel
• run the code snippet above
  --> success

But continue with:

• pip install rich
• run the snippet again
  --> failure of display.