`plot_model` `rankdir="LR"` show empty image

[khteh]

tensorflow/tensorflow#102880

Issue type

Bug

Have you reproduced the bug with TensorFlow Nightly?

No

Source

binary

TensorFlow version

2.20.0

Custom code

Yes

OS platform and distribution

Ubuntu 25.04

Mobile device

No response

Python version

3.13.3

Bazel version

No response

GCC/compiler version

No response

CUDA/cuDNN version

No response

GPU model and memory

No response

Current behavior?

Plotting a UNet vertically works perfectly but the image is too long to fit in a canocical wide-screen monitor. Plotting it horizontally makes more sense and is easier to view the correspoding skip connections. However, it doesn't work at all. The plot (saved) is empty.

Standalone code to reproduce the issue

plot_model(
    self._model,
    to_file="output/UNet.png",
    show_shapes=True,
    show_dtype=True,
    show_layer_names=True,
    rankdir="LR",
    expand_nested=True,
    show_layer_activations=True)

[dhantule]

Hi @khteh,

I tested plot_model function with rankdir="LR" (Left to Right) in this colab gist and I'm able to get the plot horizontally,
check if your model is built before plotting, provide a minimal reproducible colab if you're still facing issues.

[khteh]

Take a look at this plot. How easy is it to view it? And how "U" is it really UNet?