Model Explorer: intuitive and hierarchical visualization of model graphs

I normally use Netron for quickly inspecting models and making this &#x27;mental picture&#x27; of the architecture, but this hierarchical approach seems a better fit for my needs.<p>I&#x27;m just starting and the first impression is pretty good!

These tools are eye candy and have been around from tensorflow&#x2F;tensorboard 0.x 10 years ago but never used after just trying them for fun. You need to read the source code no easy way around it.

Looks like they have moved to <a href="https:&#x2F;&#x2F;ai.google.dev&#x2F;edge&#x2F;model-explorer" rel="nofollow">https:&#x2F;&#x2F;ai.google.dev&#x2F;edge&#x2F;model-explorer</a>

<a href="https:&#x2F;&#x2F;github.com&#x2F;lutzroeder&#x2F;netron">https:&#x2F;&#x2F;github.com&#x2F;lutzroeder&#x2F;netron</a><p><a href="https:&#x2F;&#x2F;netron.app&#x2F;" rel="nofollow">https:&#x2F;&#x2F;netron.app&#x2F;</a>

Looks cool but seems like it doesn&#x27;t work on torch 2.0<p>&quot;AttributeError: module &#x27;torch&#x27; has no attribute &#x27;export&#x27;&quot;<p>The torch.export API is currently in active development with planned breaking changes. The installation guide for this is still very minimal, anyone knows how to get it working on torch 2.0?

google-ai-edge&#x2F;model-explorer&#x2F;&#x2F;example_colabs&#x2F;quick_start.ipynb: <a href="https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;blob&#x2F;main&#x2F;example_colabs&#x2F;quick_start.ipynb">https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;blob&#x2F;main&#x2F;e...</a><p>XAI: Explainable AI: <a href="https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Explainable_artificial_intelligence" rel="nofollow">https:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Explainable_artificial_intelli...</a>

Is there a demo of a model visualized using this somewhere? Even if it&#x27;s just a short video... it&#x27;s hard to tell what it&#x27;s like from screenshots.

I&#x27;ve never really understood the point of these visualizer things. The idea that a model is always well represented by a directed acyclic graph seems extremely dated.<p>I really would love a PyTorch&#x2F;JAX profiler that shows, in annotated Python, where your code is allocating memory, using compute or doing device copies.

Other than visualizing, creating custom nodes seems to be the most interesting available operation (at the time of writing this).<p>API Guide:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;wiki&#x2F;4.-API-Guide">https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;wiki&#x2F;4.-API...</a><p>Custom Nodes - User Guide:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;wiki&#x2F;2.-User-Guide#custom-node-data">https:&#x2F;&#x2F;github.com&#x2F;google-ai-edge&#x2F;model-explorer&#x2F;wiki&#x2F;2.-Use...</a>

Is it becoming more and more common to launch open software like this without accepting contributions?

I&#x27;m confused - so it runs only in Edge or what&#x27;s with the name?!

Not to be confused with Microsoft Edge, which replaced Internet Explorer with Chromium (the browser that Google Chrome is based on). &#x2F;s

What is the &quot;AI&quot; part of this...?