# ONNX Simplifier
[](https://pypi.python.org/pypi/onnx-simplifier/)
[](https://pypi.python.org/pypi/onnx-simplifier/)
[](https://pypi.python.org/pypi/onnx-simplifier/)
[](https://github.com/daquexian/onnx-simplifier/pulls)
_ONNX is great, but sometimes too complicated._
## Background
One day I wanted to export the following simple reshape operation to ONNX:
```python
import torch
class JustReshape(torch.nn.Module):
def __init__(self):
super(JustReshape, self).__init__()
def forward(self, x):
return x.view((x.shape[0], x.shape[1], x.shape[3], x.shape[2]))
net = JustReshape()
model_name = 'just_reshape.onnx'
dummy_input = torch.randn(2, 3, 4, 5)
torch.onnx.export(net, dummy_input, model_name, input_names=['input'], output_names=['output'])
```
The input shape in this model is static, so what I expected is
However, I got the following complicated model instead:
## Our solution
ONNX Simplifier is presented to simplify the ONNX model. It infers the whole computation graph
and then replaces the redundant operators with their constant outputs (a.k.a. constant folding).
### Web version
We have published ONNX Simplifier on [convertmodel.com](https://www.convertmodel.com/#input=onnx&output=onnx). It works out of the box and **doesn't need any installation**. Note that it runs in the browser locally and your model is completely safe.
### Python version
```
pip3 install -U pip && pip3 install onnxsim
```
Then
```
onnxsim input_onnx_model output_onnx_model
```
For more advanced features, try the following command for help message
```
onnxsim -h
```
## Demonstration
An overall comparison between
[a complicated model](https://github.com/JDAI-CV/DNNLibrary/issues/17#issuecomment-455934190)
and its simplified version:
## In-script workflow
If you would like to embed ONNX simplifier python package in another script, it is just that simple.
```python
import onnx
from onnxsim import simplify
# load your predefined ONNX model
model = onnx.load(filename)
# convert model
model_simp, check = simplify(model)
assert check, "Simplified ONNX model could not be validated"
# use model_simp as a standard ONNX model object
```
You can see more details of the API in [onnxsim/onnx_simplifier.py](onnxsim/onnx_simplifier.py)
## Projects Using ONNX Simplifier
* [MXNet](https://mxnet.apache.org/versions/1.9.1/api/python/docs/tutorials/deploy/export/onnx.html#Simplify-the-exported-ONNX-model)
* [MMDetection](https://github.com/open-mmlab/mmdetection)
* [YOLOv5](https://github.com/ultralytics/yolov5)
* [ncnn](https://github.com/Tencent/ncnn)
* ...
## Chat
We created a Chinese QQ group for ONNX!
ONNX QQ Group (Chinese): 1021964010, verification code: nndab. Welcome to join!
For English users, I'm active on the [ONNX Slack](https://github.com/onnx/onnx#discuss). You can find and chat with me (daquexian) there.
Raw data
{
"_id": null,
"home_page": "https://github.com/daquexian/onnx-simplifier",
"name": "onnxsim",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.7",
"maintainer_email": "",
"keywords": "deep-learning ONNX",
"author": "ONNX Simplifier Authors",
"author_email": "daquexian566@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/ce/9e/f34238413ebeda9a3a8802feeaa5013934455466b9ab390b48ad9c7e184f/onnxsim-0.4.36.tar.gz",
"platform": null,
"description": "# ONNX Simplifier\n\n[](https://pypi.python.org/pypi/onnx-simplifier/)\n[](https://pypi.python.org/pypi/onnx-simplifier/)\n[](https://pypi.python.org/pypi/onnx-simplifier/)\n[](https://github.com/daquexian/onnx-simplifier/pulls)\n\n_ONNX is great, but sometimes too complicated._\n\n## Background\n\nOne day I wanted to export the following simple reshape operation to ONNX:\n\n```python\nimport torch\n\n\nclass JustReshape(torch.nn.Module):\n def __init__(self):\n super(JustReshape, self).__init__()\n\n def forward(self, x):\n return x.view((x.shape[0], x.shape[1], x.shape[3], x.shape[2]))\n\n\nnet = JustReshape()\nmodel_name = 'just_reshape.onnx'\ndummy_input = torch.randn(2, 3, 4, 5)\ntorch.onnx.export(net, dummy_input, model_name, input_names=['input'], output_names=['output'])\n```\n\nThe input shape in this model is static, so what I expected is\n\n\n\nHowever, I got the following complicated model instead:\n\n\n\n## Our solution\n\nONNX Simplifier is presented to simplify the ONNX model. It infers the whole computation graph\nand then replaces the redundant operators with their constant outputs (a.k.a. constant folding).\n\n### Web version\n\nWe have published ONNX Simplifier on [convertmodel.com](https://www.convertmodel.com/#input=onnx&output=onnx). It works out of the box and **doesn't need any installation**. Note that it runs in the browser locally and your model is completely safe.\n\n### Python version\n\n\n```\npip3 install -U pip && pip3 install onnxsim\n```\n\nThen\n\n```\nonnxsim input_onnx_model output_onnx_model\n```\n\nFor more advanced features, try the following command for help message\n\n```\nonnxsim -h\n```\n\n## Demonstration\n\nAn overall comparison between\n[a complicated model](https://github.com/JDAI-CV/DNNLibrary/issues/17#issuecomment-455934190)\nand its simplified version:\n\n\n\n## In-script workflow\n\nIf you would like to embed ONNX simplifier python package in another script, it is just that simple.\n\n```python\nimport onnx\nfrom onnxsim import simplify\n\n# load your predefined ONNX model\nmodel = onnx.load(filename)\n\n# convert model\nmodel_simp, check = simplify(model)\n\nassert check, \"Simplified ONNX model could not be validated\"\n\n# use model_simp as a standard ONNX model object\n```\n\nYou can see more details of the API in [onnxsim/onnx_simplifier.py](onnxsim/onnx_simplifier.py)\n\n## Projects Using ONNX Simplifier\n\n* [MXNet](https://mxnet.apache.org/versions/1.9.1/api/python/docs/tutorials/deploy/export/onnx.html#Simplify-the-exported-ONNX-model)\n* [MMDetection](https://github.com/open-mmlab/mmdetection)\n* [YOLOv5](https://github.com/ultralytics/yolov5)\n* [ncnn](https://github.com/Tencent/ncnn)\n* ...\n\n## Chat\n\nWe created a Chinese QQ group for ONNX!\n\nONNX QQ Group (Chinese): 1021964010, verification code: nndab. Welcome to join!\n\nFor English users, I'm active on the [ONNX Slack](https://github.com/onnx/onnx#discuss). You can find and chat with me (daquexian) there.\n",
"bugtrack_url": null,
"license": "Apache License v2.0",
"summary": "Simplify your ONNX model",
"version": "0.4.36",
"project_urls": {
"Homepage": "https://github.com/daquexian/onnx-simplifier"
},
"split_keywords": [
"deep-learning",
"onnx"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "9b550a0a248636cccccb7d4ed9189446e00017f411d0e13d1dd4af419ee4d529",
"md5": "b576df1aa87a40d1ba980ae23c2b5c19",
"sha256": "7498e7b9584c4b354b455564dfba66d460ce2c205b71dae169cfa9b6704e03fd"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp310-cp310-macosx_10_15_universal2.whl",
"has_sig": false,
"md5_digest": "b576df1aa87a40d1ba980ae23c2b5c19",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.7",
"size": 3471278,
"upload_time": "2024-03-04T08:24:26",
"upload_time_iso_8601": "2024-03-04T08:24:26.988344Z",
"url": "https://files.pythonhosted.org/packages/9b/55/0a0a248636cccccb7d4ed9189446e00017f411d0e13d1dd4af419ee4d529/onnxsim-0.4.36-cp310-cp310-macosx_10_15_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "d96e80c77b5c6ec079994295e6e685097fa42732a1e7c5a22fe9c5c4ca1aac74",
"md5": "4be351e6160a75e303940aa1a6f2eeaa",
"sha256": "ce87837f8975beebdcc98cc01d6d13e84b10900eb2c14035ce1066c3d670d96d"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "4be351e6160a75e303940aa1a6f2eeaa",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.7",
"size": 2255237,
"upload_time": "2024-03-04T08:24:29",
"upload_time_iso_8601": "2024-03-04T08:24:29.047750Z",
"url": "https://files.pythonhosted.org/packages/d9/6e/80c77b5c6ec079994295e6e685097fa42732a1e7c5a22fe9c5c4ca1aac74/onnxsim-0.4.36-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "0ec56c93b354684b3fc4b520a23be3e4db5870b35dde9e9e2a1f41018ba369e8",
"md5": "caf0f4c427d4d5ccf3e171b4ae4a40c0",
"sha256": "f92bec8c6c0d4f8463e10021277711d2faac900e4eb890238001b3eadb5c03bc"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp310-cp310-win_amd64.whl",
"has_sig": false,
"md5_digest": "caf0f4c427d4d5ccf3e171b4ae4a40c0",
"packagetype": "bdist_wheel",
"python_version": "cp310",
"requires_python": ">=3.7",
"size": 1288644,
"upload_time": "2024-03-04T08:24:31",
"upload_time_iso_8601": "2024-03-04T08:24:31.275869Z",
"url": "https://files.pythonhosted.org/packages/0e/c5/6c93b354684b3fc4b520a23be3e4db5870b35dde9e9e2a1f41018ba369e8/onnxsim-0.4.36-cp310-cp310-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "2ea3f6cad8499f375a3acc8a7837721f82860244656cf62984cf80ebe187cc68",
"md5": "c29864f09ee90bf93413f14405982a1c",
"sha256": "150b9a3a409af2f3161af3fecda2113e0e6e296fb015b5205a9ddf645765acad"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp311-cp311-macosx_10_15_universal2.whl",
"has_sig": false,
"md5_digest": "c29864f09ee90bf93413f14405982a1c",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.7",
"size": 3471242,
"upload_time": "2024-03-04T08:24:33",
"upload_time_iso_8601": "2024-03-04T08:24:33.289473Z",
"url": "https://files.pythonhosted.org/packages/2e/a3/f6cad8499f375a3acc8a7837721f82860244656cf62984cf80ebe187cc68/onnxsim-0.4.36-cp311-cp311-macosx_10_15_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "db9422aab761b3d416bce02020d9ca98dc692427c2717b0325952e30ce41f83b",
"md5": "1e24ddc01e915354be6077879919eeb9",
"sha256": "fa7596e6b806ed19077f7652788a50ee576c172b4d16d421f0593aef1a6fa4c4"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "1e24ddc01e915354be6077879919eeb9",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.7",
"size": 2255003,
"upload_time": "2024-03-04T08:24:35",
"upload_time_iso_8601": "2024-03-04T08:24:35.024038Z",
"url": "https://files.pythonhosted.org/packages/db/94/22aab761b3d416bce02020d9ca98dc692427c2717b0325952e30ce41f83b/onnxsim-0.4.36-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "c15caa277f45b0d8253027d1ce3269952e116b476985e5fb497e00ebd917ce29",
"md5": "30555dbc99267f6c30d2fe84c9540deb",
"sha256": "91fb32def04f2f89d5f76527c852332366957752e5e61ac25be0b2d7bb410f89"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp311-cp311-win_amd64.whl",
"has_sig": false,
"md5_digest": "30555dbc99267f6c30d2fe84c9540deb",
"packagetype": "bdist_wheel",
"python_version": "cp311",
"requires_python": ">=3.7",
"size": 1288684,
"upload_time": "2024-03-04T08:24:37",
"upload_time_iso_8601": "2024-03-04T08:24:37.064924Z",
"url": "https://files.pythonhosted.org/packages/c1/5c/aa277f45b0d8253027d1ce3269952e116b476985e5fb497e00ebd917ce29/onnxsim-0.4.36-cp311-cp311-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "03c0e64f4d92bee6596620417f1d388bee7a73ecc8e0caf173c844d7bd8e22d0",
"md5": "40e3e2fe494e26101d6e2b246d5d312f",
"sha256": "9c616b1e48e92d916e762bae405563f5b2fd8d0718fbbafef2ceb81570202716"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "40e3e2fe494e26101d6e2b246d5d312f",
"packagetype": "bdist_wheel",
"python_version": "cp37",
"requires_python": ">=3.7",
"size": 2258345,
"upload_time": "2024-03-04T08:24:39",
"upload_time_iso_8601": "2024-03-04T08:24:39.535904Z",
"url": "https://files.pythonhosted.org/packages/03/c0/e64f4d92bee6596620417f1d388bee7a73ecc8e0caf173c844d7bd8e22d0/onnxsim-0.4.36-cp37-cp37m-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e3c079fc142a53a27e477f0f82d2311600e343a6da700e5407af03dc389ce8b9",
"md5": "479a83fc4e3fcc6340c68e5c9dea8b84",
"sha256": "7c0aa3de159fc9c4560b62ddccc0d1146fef035aeba5eb3797f6550889a2f96a"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp37-cp37m-win_amd64.whl",
"has_sig": false,
"md5_digest": "479a83fc4e3fcc6340c68e5c9dea8b84",
"packagetype": "bdist_wheel",
"python_version": "cp37",
"requires_python": ">=3.7",
"size": 1288818,
"upload_time": "2024-03-04T08:24:41",
"upload_time_iso_8601": "2024-03-04T08:24:41.339897Z",
"url": "https://files.pythonhosted.org/packages/e3/c0/79fc142a53a27e477f0f82d2311600e343a6da700e5407af03dc389ce8b9/onnxsim-0.4.36-cp37-cp37m-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "3a933b27939bcbde20d04088770b6cda2e0c4211335732329aa7cd36f78f89ca",
"md5": "0c4560a86d88c9c29b89b0cd66effd52",
"sha256": "369815cd292ae1b3cb37d5692428abf69ed3917c7794d895155c4d4d1f8b374e"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp38-cp38-macosx_10_15_universal2.whl",
"has_sig": false,
"md5_digest": "0c4560a86d88c9c29b89b0cd66effd52",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.7",
"size": 3465984,
"upload_time": "2024-03-04T08:24:43",
"upload_time_iso_8601": "2024-03-04T08:24:43.299206Z",
"url": "https://files.pythonhosted.org/packages/3a/93/3b27939bcbde20d04088770b6cda2e0c4211335732329aa7cd36f78f89ca/onnxsim-0.4.36-cp38-cp38-macosx_10_15_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "30e625dd3bd1dbb30561662d8c2e9e5aa014a2e704d1a81053fe9e935d5fe6be",
"md5": "82f288db317ef799fd9eed2f51bb4437",
"sha256": "19c881a9c9af4e7238c856da8b677e158f9a74f3eeaedbdb447fb109a4d08ce3"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "82f288db317ef799fd9eed2f51bb4437",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.7",
"size": 2254975,
"upload_time": "2024-03-04T08:24:44",
"upload_time_iso_8601": "2024-03-04T08:24:44.797964Z",
"url": "https://files.pythonhosted.org/packages/30/e6/25dd3bd1dbb30561662d8c2e9e5aa014a2e704d1a81053fe9e935d5fe6be/onnxsim-0.4.36-cp38-cp38-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "2cbc6f4248187a13cec1782854d4fb339c32d8976821fb74694d809f7a4bb3eb",
"md5": "919c464a846627b36e333f7c13b32030",
"sha256": "f17c2b81ad97e7bb91f6f049282b7d9bc8e4dc93d23d2bdf59e8bf3f5b3e63c7"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp38-cp38-win_amd64.whl",
"has_sig": false,
"md5_digest": "919c464a846627b36e333f7c13b32030",
"packagetype": "bdist_wheel",
"python_version": "cp38",
"requires_python": ">=3.7",
"size": 1288766,
"upload_time": "2024-03-04T08:24:47",
"upload_time_iso_8601": "2024-03-04T08:24:47.113612Z",
"url": "https://files.pythonhosted.org/packages/2c/bc/6f4248187a13cec1782854d4fb339c32d8976821fb74694d809f7a4bb3eb/onnxsim-0.4.36-cp38-cp38-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "203b2230a57ee0425649b16bf8ba0a17400c9aa003f049052833055ff3bcbfa8",
"md5": "5f0ca276d57517ecb47867e9315ddf2f",
"sha256": "6ae38cb9a23b759202d8ecc6c25392a2cecd156499b036fc48720d048833bece"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp39-cp39-macosx_10_15_universal2.whl",
"has_sig": false,
"md5_digest": "5f0ca276d57517ecb47867e9315ddf2f",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.7",
"size": 3471439,
"upload_time": "2024-03-04T08:24:48",
"upload_time_iso_8601": "2024-03-04T08:24:48.950503Z",
"url": "https://files.pythonhosted.org/packages/20/3b/2230a57ee0425649b16bf8ba0a17400c9aa003f049052833055ff3bcbfa8/onnxsim-0.4.36-cp39-cp39-macosx_10_15_universal2.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "0cc321c4e4a17c0dae64c0f0c02186baa198d7d5563af9a27361d2fdd75a8e04",
"md5": "ac25e5b1299802f28ab7729dccf2e5be",
"sha256": "e937abb8e20a6609f27ae19639d21dc5e8621c4a5e44ebbafab9292451f75497"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"has_sig": false,
"md5_digest": "ac25e5b1299802f28ab7729dccf2e5be",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.7",
"size": 2255609,
"upload_time": "2024-03-04T08:24:50",
"upload_time_iso_8601": "2024-03-04T08:24:50.594707Z",
"url": "https://files.pythonhosted.org/packages/0c/c3/21c4e4a17c0dae64c0f0c02186baa198d7d5563af9a27361d2fdd75a8e04/onnxsim-0.4.36-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "adc416260e99ed9f82e33d4061a953d927bd174e0c67868e4e15ae5335cd1713",
"md5": "5703fe9086804fb4a025b99840b39664",
"sha256": "1a23f32bc6ab3829ee69d9b737e4cfeb457d91720c4040788dabbc88e11dfacc"
},
"downloads": -1,
"filename": "onnxsim-0.4.36-cp39-cp39-win_amd64.whl",
"has_sig": false,
"md5_digest": "5703fe9086804fb4a025b99840b39664",
"packagetype": "bdist_wheel",
"python_version": "cp39",
"requires_python": ">=3.7",
"size": 1288184,
"upload_time": "2024-03-04T08:24:52",
"upload_time_iso_8601": "2024-03-04T08:24:52.271866Z",
"url": "https://files.pythonhosted.org/packages/ad/c4/16260e99ed9f82e33d4061a953d927bd174e0c67868e4e15ae5335cd1713/onnxsim-0.4.36-cp39-cp39-win_amd64.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "ce9ef34238413ebeda9a3a8802feeaa5013934455466b9ab390b48ad9c7e184f",
"md5": "9082117772abf4598364d63e59c4672d",
"sha256": "6e0ee9d6d4a83042bdef7319fbe58352d9fda5f253386be2b267c7c27f0638ee"
},
"downloads": -1,
"filename": "onnxsim-0.4.36.tar.gz",
"has_sig": false,
"md5_digest": "9082117772abf4598364d63e59c4672d",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.7",
"size": 20993703,
"upload_time": "2024-03-04T08:25:00",
"upload_time_iso_8601": "2024-03-04T08:25:00.086854Z",
"url": "https://files.pythonhosted.org/packages/ce/9e/f34238413ebeda9a3a8802feeaa5013934455466b9ab390b48ad9c7e184f/onnxsim-0.4.36.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2024-03-04 08:25:00",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "daquexian",
"github_project": "onnx-simplifier",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"requirements": [],
"lcname": "onnxsim"
}
JSON
