Skip to main content

AutoGrad Visualizer (agviz): Visualize dependencies between states/parameters in autograd tree.

Project description

agviz

AutoGrad Visualizer (agviz): Visualize dependencies between states/parameters in autograd tree.

Unlike related packages (e.g. torvhviz), supports visualizing internal model states and distinguishing states/parameters. Also supports more complex visualization based on traversals of graph and merging nodes.

Installation:

In the root directory run pip install -e .
You can then install the library in your python code using `from adviz import render' (see Examples)

Examples:

examples/gru: Render the hidden states of a GRU for 5 timesteps, visualizing parameters and states.

Simple GRU Example


examples/rnn: Render the hidden states of a RNN for 5 timesteps, visualizing parameters and states. Also, play with "full" rendering, where we show all the intermediate functions.

Simple RNN Example

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

agviz-1.0.1.tar.gz (5.2 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

agviz-1.0.1-py3-none-any.whl (6.0 kB view details)

Uploaded Python 3

File details

Details for the file agviz-1.0.1.tar.gz.

File metadata

  • Download URL: agviz-1.0.1.tar.gz
  • Upload date:
  • Size: 5.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for agviz-1.0.1.tar.gz
Algorithm Hash digest
SHA256 94f112516431e389b1d0ceb7162b3955311bd7e07477211dc60f3f0a0c3af0ef
MD5 99740c2bd32b2fbc75b25b9803886a95
BLAKE2b-256 a2dc493214bb00ba62eabff13c5770806e017dd7d3e74fbb9b47e20167371357

See more details on using hashes here.

File details

Details for the file agviz-1.0.1-py3-none-any.whl.

File metadata

  • Download URL: agviz-1.0.1-py3-none-any.whl
  • Upload date:
  • Size: 6.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.9

File hashes

Hashes for agviz-1.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 49353386b1a0792fa24faa99709838d15b7b6720107c7759fd2dfb1bab9fb211
MD5 5f8aaa5015384cd51f21180a7ee049dd
BLAKE2b-256 0b0b52cab06769eb93c1fcc81ac67f00fe80073ba45ca50d51fd32727afc036b

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page