A visualization tool to show a TensorFlow's graph like TensorBoard

Last update: Nov 09, 2022

Overview

tfgraphviz

tfgraphviz is a module to visualize a TensorFlow's data flow graph like TensorBoard using Graphviz. tfgraphviz enables to provide a visualization of tensorflow graph on Jupyter Notebook without TensorBoard.

Installation

Use pip to install:

$ pip install graphviz
$ pip install tfgraphviz

The only dependency is Graphviz.

macOS:

$ brew install graphviz

Ubuntu:

$ apt-get install graphviz

Quickstart

import tensorflow as tf
import tfgraphviz as tfg

g = tf.Graph()
with g.as_default():
    a = tf.constant(1, name="a")
    b = tf.constant(2, name="b")
    c = tf.add(a, b, name="add")
tfg.board(g)

License

This package is distributed under the MIT license.

Author

Akimasa KIMURA

Comments

Add Binder support

To get the example working in Binder the only required pip installable package is tensorflow and from apt-get is graphviz. The postBuild will install tfgraphviz from GitHub so that it always gets the master HEAD, so that if there is something broken for a bit on PyPI it won't continue to be a problem until a new release is made.

opened by matthewfeickert 2
added pan and zoom support for jupyter
Added tfg.jupyter_pan_and_zoom helper to wrap generated SVG object In order to enable pan and zoom functionality in Jupyter:

tfg.jupyter_pan_and_zoom(tfg.board(graph))

Tested in JupyterLab and Colab.
opened by vlasenkoalexey 1

Extending functionality

Added proper tooltips
Fixed logic to render function names to check op type, not op name
Added ability to override functions to create digraph, node and edge like:

def custom_add_digraph_node(digraph, name, op, attributes=None):
    attributes=[]
    if op is not None and 'PartitionedCall' in op.type:
        attributes.append(('fillcolor', 'blue'))
    tfg.add_digraph_node(digraph, name, op, attributes)

tfg.board(tf_g, depth=10, name_regex=".*", add_digraph_node_func=custom_add_digraph_node)

opened by vlasenkoalexey 1

For Python3.x
Fixed to work on Python 3.x:

Change IMPLICIT relative imports (from graphviz_wrapper import board) to EXPLICIT relative imports (from .graphviz_wrapper import board).

(In addition, ) remove unnecessary imports.

Checked to work both Python 2.7.x/3.5.x with TensorFlow 1.0.
opened by antimon2 1

UnicodeDecodeError on loading the graph.

I am getting the following error. Is unicode in tf variable/scope names not supported?

<ipython-input-2-b2099ef84663> in load(self)
    143         # sess = tf.Session(graph=tf.get_default_graph())
    144         self.sess.run(self.init)
--> 145         g = tfg.board(tf.get_default_graph())
    146         g.view()
    147         self.saver.restore(self.sess, model_path)

/Users/activeai/.local/share/virtualenvs/spotter-67LXOL6z/lib/python2.7/site-packages/tfgraphviz/graphviz_wrapper.pyc in board(tfgraph, depth, name, style)
    212     _node_inpt_table, _node_inpt_shape_table = node_input_table(tfgraph, depth=depth)
    213     digraph = add_nodes(_node_table, name=name, style=style)
--> 214     digraph = add_edges(digraph, _node_inpt_table, _node_inpt_shape_table)
    215     return digraph

/Users/activeai/.local/share/virtualenvs/spotter-67LXOL6z/lib/python2.7/site-packages/tfgraphviz/graphviz_wrapper.pyc in add_edges(digraph, node_inpt_table, node_inpt_shape_table)
    195             else:
    196                 shape = node_inpt_shape_table[ni]
--> 197                 digraph.edge(ni, node, label=edge_label(shape))
    198     return digraph
    199 

/Users/activeai/.local/share/virtualenvs/spotter-67LXOL6z/lib/python2.7/site-packages/graphviz/dot.pyc in edge(self, tail_name, head_name, label, _attributes, **attrs)
    145         head_name = self._quote_edge(head_name)
    146         attr_list = self._attr_list(label, attrs, _attributes)
--> 147         line = self._edge % (tail_name, head_name, attr_list)
    148         self.body.append(line)
    149 

UnicodeDecodeError: 'ascii' codec can't decode byte 0xc3 in position 10: ordinal not in range(128)

opened by malarinv 0

Releases(0.0.8)

0.0.8(Mar 6, 2021)

#7 and #8 are Merged.
Source code(tar.gz)
Source code(zip)
0.0.7(Jan 8, 2021)

Source code(tar.gz)
Source code(zip)
0.0.6(May 30, 2017)

Source code(tar.gz)
Source code(zip)
0.0.5(Mar 3, 2017)

Source code(tar.gz)
Source code(zip)
0.0.4(Feb 20, 2017)

Source code(tar.gz)
Source code(zip)
0.0.3(Feb 18, 2017)

Source code(tar.gz)
Source code(zip)
0.0.0(Feb 16, 2017)

Source code(tar.gz)
Source code(zip)

Owner

GitHub Repository https://pypi.python.org/pypi/tfgraphviz

This is an official pytorch implementation of Lite-HRNet: A Lightweight High-Resolution Network.

Lite-HRNet: A Lightweight High-Resolution Network Introduction This is an official pytorch implementation of Lite-HRNet: A Lightweight High-Resolution

675 Dec 25, 2022

image scene graph generation benchmark

Scene Graph Benchmark in PyTorch 1.7 This project is based on maskrcnn-benchmark Highlights Upgrad to pytorch 1.7 Multi-GPU training and inference Bat

303 Dec 27, 2022

A developer interface for creating Chat AIs for the Chai app.

ChaiPy A developer interface for creating Chat AIs for the Chai app. Usage Local development A quick start guide is available here, with a minimal exa

28 Dec 28, 2022

SwinIR: Image Restoration Using Swin Transformer

SwinIR: Image Restoration Using Swin Transformer This repository is the official PyTorch implementation of SwinIR: Image Restoration Using Shifted Win

2.4k Jan 05, 2023

Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning (ICLR 2021)

Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning (ICLR 2021) Citation Please cite as: @inproceedings{liu2020understan

22 Nov 25, 2022

scAR (single-cell Ambient Remover) is a package for data denoising in single-cell omics.

scAR scAR (single cell Ambient Remover) is a package for denoising multiple single cell omics data. It can be used for multiple tasks, such as, sgRNA

19 Nov 28, 2022

A framework to train language models to learn invariant representations.

Invariant Language Modeling Implementation of the training for invariant language models. Motivation Modern pretrained language models are critical co

6 Nov 16, 2022

Implementation of OmniNet, Omnidirectional Representations from Transformers, in Pytorch

Omninet - Pytorch Implementation of OmniNet, Omnidirectional Representations from Transformers, in Pytorch. The authors propose that we should be atte

48 Nov 21, 2022

YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone

YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone In our recent paper we propose the YourTTS model. YourTTS bri

390 Dec 29, 2022

Enabling dynamic analysis of Legacy Embedded Systems in full emulated environment

PENecro This project is based on "Enabling dynamic analysis of Legacy Embedded Systems in full emulated environment", published on hardwear.io USA 202

10 May 17, 2022

JupyterNotebook - C/C++, Javascript, HTML, LaTex, Shell scripts in Jupyter Notebook Also run them on remote computer

JupyterNotebook Read, write and execute C, C++, Javascript, Shell scripts, HTML, LaTex in jupyter notebook, And also execute them on remote computer R

1 Jan 09, 2022

Vignette is a face tracking software for characters using osu!framework.

Vignette is a face tracking software for characters using osu!framework. Unlike most solutions, Vignette is: Made with osu!framework, the game framewo

412 Dec 28, 2022

Unofficial pytorch implementation of the paper "Dynamic High-Pass Filtering and Multi-Spectral Attention for Image Super-Resolution"

DFSA Unofficial pytorch implementation of the ICCV 2021 paper "Dynamic High-Pass Filtering and Multi-Spectral Attention for Image Super-Resolution" (p

2 Nov 15, 2021

A visualization tool to show a TensorFlow's graph like TensorBoard

Related tags

Overview

tfgraphviz

Links

Installation

Quickstart

License

Author

Comments

Add Binder support

added pan and zoom support for jupyter

Extending functionality

For Python3.x

UnicodeDecodeError on loading the graph.

Releases(0.0.8)

0.0.8(Mar 6, 2021)

0.0.7(Jan 8, 2021)

0.0.6(May 30, 2017)

0.0.5(Mar 3, 2017)

0.0.4(Feb 20, 2017)

0.0.3(Feb 18, 2017)

0.0.0(Feb 16, 2017)

Owner

This is an official pytorch implementation of Lite-HRNet: A Lightweight High-Resolution Network.

image scene graph generation benchmark

A developer interface for creating Chat AIs for the Chai app.

SwinIR: Image Restoration Using Swin Transformer

Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning (ICLR 2021)

scAR (single-cell Ambient Remover) is a package for data denoising in single-cell omics.

A framework to train language models to learn invariant representations.

Implementation of OmniNet, Omnidirectional Representations from Transformers, in Pytorch

YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone

Enabling dynamic analysis of Legacy Embedded Systems in full emulated environment

JupyterNotebook - C/C++, Javascript, HTML, LaTex, Shell scripts in Jupyter Notebook Also run them on remote computer

Vignette is a face tracking software for characters using osu!framework.

Unofficial pytorch implementation of the paper "Dynamic High-Pass Filtering and Multi-Spectral Attention for Image Super-Resolution"

ShuttleNet: Position-aware Fusion of Rally Progress and Player Styles for Stroke Forecasting in Badminton (AAAI'22)

Search Youtube Video and Get Video info

A Lighting Pytorch Framework for Recommendation System, Easy-to-use and Easy-to-extend.

Implementation of Google Brain's WaveGrad high-fidelity vocoder

Implementation of ML models like Decision tree, Naive Bayes, Logistic Regression and many other

Conformer: Local Features Coupling Global Representations for Visual Recognition

Count the MACs / FLOPs of your PyTorch model.