SunnyMirror/netron @ d47afddd4ff9319391f4b7bc0efe58c4aa18e38c

Нема описа

Lutz Roeder d47afddd4f Add MXNet metadata		пре 8 година
.vscode	1a3875886e first commit	пре 8 година
media	c58cf0a220 Update screenshot	пре 8 година
setup	e4e4bc7743 Move electron-builder setup files to /setup	пре 8 година
src	d47afddd4f Add MXNet metadata	пре 8 година
tools	04e5563839 MXNet symbol.json support	пре 8 година
.gitattributes	7f59d7662f Support variadic and control edges	пре 8 година
.gitignore	aef727bc5d Add src/coreml.js	пре 8 година
DEVELOPMENT.md	b2da722b5c Update DEVELOPMENT.md	пре 8 година
LICENSE	0157c86a27 Update license	пре 9 година
Makefile	3fc2eef9df Delay load protobuf and library scripts	пре 8 година
README.md	04e5563839 MXNet symbol.json support	пре 8 година
electron-builder.yml	04e5563839 MXNet symbol.json support	пре 8 година
netron	eaf15f719b Fix Python scripts for 3.x	пре 8 година
package.json	d5a70ca06b Update to 1.6.1	пре 8 година
setup.cfg	bb17ed372c All Python build outputs to ./build/python	пре 8 година
setup.py	04e5563839 MXNet symbol.json support	пре 8 година

Netron is a viewer for neural network and machine learning models.

Netron supports ONNX (.onnx, .pb), Keras (.h5, .keras), CoreML (.mlmodel) and TensorFlow Lite (.tflite).

Netron has experimental support for Caffe (.caffemodel), MXNet (-symbol.json) and TensorFlow (.pb, .meta).

Install

macOS

Download the .dmg file or with Homebrew run brew cask install netron

Linux

Download the .AppImage or .deb file.

Windows

Download the .exe installer.

Browser

Start the browser version.

Python Server

Run pip install netron and netron [MODEL_FILE].
Serve a model in Python using import netron; netron.serve_file('my_model.onnx').

Sample model files you can download and open:

ONNX Models

Keras Models

CoreML Models

TensorFlow Lite Models

Caffe Models

TensorFlow models