Simple node deletion tool for onnx.

Last update: May 15, 2022

Overview

snd4onnx

Simple node deletion tool for onnx. I only test very miscellaneous and limited patterns as a hobby. There are probably a large number of bugs. Pull requests are welcome.

https://github.com/PINTO0309/simple-onnx-processing-tools

1. Setup

1-1. HostPC

### option
$ echo export PATH="~/.local/bin:$PATH" >> ~/.bashrc \
&& source ~/.bashrc

### run
$ pip install -U onnx \
&& python3 -m pip install -U onnx_graphsurgeon --index-url https://pypi.ngc.nvidia.com \
&& pip install -U snd4onnx

1-2. Docker

### docker pull
$ docker pull pinto0309/snd4onnx:latest

### docker build
$ docker build -t pinto0309/snd4onnx:latest .

### docker run
$ docker run --rm -it -v `pwd`:/workdir pinto0309/snd4onnx:latest
$ cd /workdir

2. CLI Usage

$ snd4onnx -h

usage:
  snd4onnx [-h]
    --remove_node_names REMOVE_NODE_NAMES [REMOVE_NODE_NAMES ...]
    --input_onnx_file_path INPUT_ONNX_FILE_PATH
    --output_onnx_file_path OUTPUT_ONNX_FILE_PATH

optional arguments:
  -h, --help
        show this help message and exit

  --remove_node_names REMOVE_NODE_NAMES [REMOVE_NODE_NAMES ...]
        ONNX node name to be deleted.

  --input_onnx_file_path INPUT_ONNX_FILE_PATH
        Input onnx file path.

  --output_onnx_file_path OUTPUT_ONNX_FILE_PATH
        Output onnx file path.

3. In-script Usage

>>> from snd4onnx import remove
>>> help(remove)

Help on function remove in module snd4onnx.onnx_remove_node:

remove(
    remove_node_names: List[str],
    input_onnx_file_path: Union[str, NoneType] = '',
    output_onnx_file_path: Union[str, NoneType] = '',
    onnx_graph: Union[onnx.onnx_ml_pb2.ModelProto, NoneType] = None
) -> onnx.onnx_ml_pb2.ModelProto

    Parameters
    ----------
    remove_node_names: List[str]
        List of OP names to be deleted.
        e.g. remove_node_names = ['op_name1', 'op_name2', 'op_name3', ...]

    input_onnx_file_path: Optional[str]
        Input onnx file path.
        Either input_onnx_file_path or onnx_graph must be specified.

    output_onnx_file_path: Optional[str]
        Output onnx file path.
        If output_onnx_file_path is not specified, no .onnx file is output.

    onnx_graph: Optional[onnx.ModelProto]
        onnx.ModelProto.
        Either input_onnx_file_path or onnx_graph must be specified.
        onnx_graph If specified, ignore input_onnx_file_path and process onnx_graph.

    Returns
    -------
    removed_graph: onnx.ModelProto
        OP removed onnx ModelProto.

4. CLI Execution

$ snd4onnx \
--remove_node_names node_name_a node_name_b
--input_onnx_file_path input.onnx \
--output_onnx_file_path output.onnx

5. In-script Execution

from snd4onnx import remove

onnx_graph = remove(
    remove_node_names=['node_name_a', 'node_name_b'],
    input_onnx_file_path='input.onnx',
)

# or

onnx_graph = remove(
    remove_node_names=['node_name_a', 'node_name_b'],
    onnx_graph=graph,
)

6. Sample

6-1. sample.1

Before	After

6-2. sample.2

Before	After

6-3. sample.3

Before	After

6-4. sample.4

Before	After

6-5. sample.5

Before	After

7. Reference

8. Issues

https://github.com/PINTO0309/simple-onnx-processing-tools/issues

You might also like...

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.

MMdnn MMdnn is a comprehensive and cross-framework tool to convert, visualize and diagnose deep learning (DL) models. The "MM" stands for model manage

5.7k Jan 9, 2023

PyTorch ,ONNX and TensorRT implementation of YOLOv4

4.2k Jan 1, 2023

YOLOv5 in PyTorch ONNX CoreML TFLite

This repository represents Ultralytics open-source research into future object detection methods, and incorporates lessons learned and best practices evolved over thousands of hours of training and evolution on anonymized client datasets. All code and models are under active development, and are subject to modification or deletion without notice.

34.1k Dec 31, 2022

Releases(1.1.6)

1.1.6(Sep 8, 2022)

Add short form parameter

$ snd4onnx -h

usage:
    snd4onnx [-h]
    -rn REMOVE_NODE_NAMES [REMOVE_NODE_NAMES ...]
    -if INPUT_ONNX_FILE_PATH
    -of OUTPUT_ONNX_FILE_PATH
    [-n]

optional arguments:
  -h, --help
        show this help message and exit.

  -rn REMOVE_NODE_NAMES [REMOVE_NODE_NAMES ...], --remove_node_names REMOVE_NODE_NAMES [REMOVE_NODE_NAMES ...]
        ONNX node name to be deleted.

  -if INPUT_ONNX_FILE_PATH, --input_onnx_file_path INPUT_ONNX_FILE_PATH
        Input onnx file path.

  -of OUTPUT_ONNX_FILE_PATH, --output_onnx_file_path OUTPUT_ONNX_FILE_PATH
        Output onnx file path.

  -n, --non_verbose
        Do not show all information logs. Only error logs are displayed.

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

1.1.5(May 25, 2022)
Security update

README update

Source code(tar.gz)
Source code(zip)
1.1.4(May 20, 2022)
https://github.com/PINTO0309/snd4onnx/discussions/2 f-string fixes

Source code(tar.gz)
Source code(zip)
1.1.3(May 19, 2022)
Support for deletion of output_nodes

Source code(tar.gz)
Source code(zip)
1.1.2(Apr 16, 2022)
Bug fixes

Source code(tar.gz)
Source code(zip)
1.1.1(Apr 11, 2022)
Support for Docker

Support for onnx.ModelProto

Source code(tar.gz)
Source code(zip)
1.1.0(Apr 10, 2022)
In-script support

Source code(tar.gz)
Source code(zip)
1.0.0(Apr 10, 2022)
README update

Source code(tar.gz)
Source code(zip)
0.0.2(Apr 1, 2022)
README update

Source code(tar.gz)
Source code(zip)
0.0.1(Apr 1, 2022)

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

Simple node deletion tool for onnx.

Related tags

Overview

snd4onnx

1. Setup

1-1. HostPC

1-2. Docker

2. CLI Usage

3. In-script Usage

4. CLI Execution

5. In-script Execution

6. Sample

6-1. sample.1

6-2. sample.2

6-3. sample.3

6-4. sample.4

6-5. sample.5

7. Reference

8. Issues

You might also like...

MMdnn is a set of tools to help users inter-operate among different deep learning frameworks. E.g. model conversion and visualization. Convert models between Caffe, Keras, MXNet, Tensorflow, CNTK, PyTorch Onnx and CoreML.

PyTorch ,ONNX and TensorRT implementation of YOLOv4

YOLOv5 in PyTorch ONNX CoreML TFLite

A code generator from ONNX to PyTorch code

tf2onnx - Convert TensorFlow, Keras and Tflite models to ONNX.

Export CenterPoint PonintPillars ONNX Model For TensorRT

A high-performance anchor-free YOLO. Exceeding yolov3~v5 with ONNX, TensorRT, NCNN, and Openvino supported.

YOLOv3 in PyTorch ONNX CoreML TFLite

YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with ONNX, TensorRT, ncnn, and OpenVINO supported.

Releases(1.1.6)

1.1.6(Sep 8, 2022)

1.1.5(May 25, 2022)

1.1.4(May 20, 2022)

1.1.3(May 19, 2022)

1.1.2(Apr 16, 2022)

1.1.1(Apr 11, 2022)

1.1.0(Apr 10, 2022)

1.0.0(Apr 10, 2022)

0.0.2(Apr 1, 2022)

0.0.1(Apr 1, 2022)

Owner

Katsuya Hyodo

Portfolio Optimization and Quantitative Strategic Asset Allocation in Python

Keras Image Embeddings using Contrastive Loss

[CVPR 2021] Counterfactual VQA: A Cause-Effect Look at Language Bias

ROS support for Velodyne 3D LIDARs

Explainable Zero-Shot Topic Extraction

Rule Based Classification Project For Python

Weak-supervised Visual Geo-localization via Attention-based Knowledge Distillation

Learning What and Where to Draw

A framework for Quantification written in Python

Source code related to the article submitted to the International Conference on Computational Science ICCS 2022 in London

Python scripts for performing stereo depth estimation using the MobileStereoNet model in Tensorflow Lite.

(Personalized) Page-Rank computation using PyTorch

Problem-943.-ACMP - Problem 943. ACMP

Official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers, a novel method to visualize any Transformer-based network. Including examples for DETR, VQA.

DEMix Layers for Modular Language Modeling

这是一个yolox-pytorch的源码，可以用于训练自己的模型。

Auto grind btdb2 exp for tower

BMN: Boundary-Matching Network

Official pytorch implementation of paper "Image-to-image Translation via Hierarchical Style Disentanglement".

Code for models used in Bashiri et al., "A Flow-based latent state generative model of neural population responses to natural images".