blocks|key|458515|text|在加载带泡菜的模型时，源树必须与保存模型时使用的树匹配。所以|type|unstyled|depth|inlineStyleRanges|entityRanges|data|458516|ModuleNotFoundError:+No+module+named+'models'|code-block|syntax|javascript|458517|期望这个目录位于您的python路径中：https://github.com/WongKinYiu/yolov7/tree/main/models|offset|length|458518|向ONNX出口：|458519|克隆回购https://github.com/WongKinYiu/yolov7|ordered-list-item|458520|git+clone+https://github.com/WongKinYiu/yolov7|458521|设置正确的路径。|458522|import+sys
sys.path.insert(0,+'./yolov7')|458523|也可以设置PYTHONPATH环境变量。|style|CODE|458524|此外，您可能需要有特定的torch版本。我查过了，看来torch==1.8.0还不错|458525|示例：|458526|import+torch
import+sys

sys.path.insert(0,+'./yolov7')

device+=+torch.device('cpu')
model+=+torch.load('yolov7x.pt',+map_location=device)['model'].float()
torch.onnx.export(model,+torch.zeros((1,+3,+640,+640)),+'yolov7.onnx',+opset_version=12)|458527|在此之后，模型被导出到ONNX+(用内特恩可视化)：|458528|📷|atomic|458529|​|458530|通常，最好将权重保存为state_dict，并保留可以重构torch.nn.Module的源代码，这样您就可以安全地使用：|458531|model.load_state_dict(torch.load('weights.pt'))|458532|entityMap|0|LINK|mutability|MUTABLE|url|https://github.com/WongKinYiu/yolov7/tree/main/models|1|https://github.com/WongKinYiu/yolov7|2|https://github.com/lutzroeder/netron|3|IMAGE|IMMUTABLE|imageUrl|https://developer.qcloudimg.com/http-save/yehe-900000/958a0bd577f6e7b591b9cb42bfa2ff7c.png|imageAlt^0|0|0|K|1H|0|0|0|4|10|1|0|0|0|0|5|A|0|C|5|R|C|0|0|0|I|3|2|0|0|1|3|0|0|B|A|T|F|0|0|1B|0^^$0|@$1|2|3|4|5|6|7|1Y|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|1Z|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|20|8|@]|9|@$I|21|J|22|1|23]]|A|$]]|$1|K|3|L|5|6|7|24|8|@]|9|@]|A|$]]|$1|M|3|N|5|O|7|25|8|@]|9|@$I|26|J|27|1|28]]|A|$]]|$1|P|3|Q|5|D|7|29|8|@]|9|@]|A|$E|F]]|$1|R|3|S|5|O|7|2A|8|@]|9|@]|A|$]]|$1|T|3|U|5|D|7|2B|8|@]|9|@]|A|$E|F]]|$1|V|3|W|5|6|7|2C|8|@$I|2D|J|2E|X|Y]]|9|@]|A|$]]|$1|Z|3|10|5|O|7|2F|8|@$I|2G|J|2H|X|Y]|$I|2I|J|2J|X|Y]]|9|@]|A|$]]|$1|11|3|12|5|6|7|2K|8|@]|9|@]|A|$]]|$1|13|3|14|5|D|7|2L|8|@]|9|@]|A|$E|F]]|$1|15|3|16|5|6|7|2M|8|@]|9|@$I|2N|J|2O|1|2P]]|A|$]]|$1|17|3|18|5|19|7|2Q|8|@]|9|@$I|2R|J|2S|1|2T]]|A|$]]|$1|1A|3|1B|5|6|7|2U|8|@]|9|@]|A|$]]|$1|1C|3|1D|5|6|7|2V|8|@$I|2W|J|2X|X|Y]|$I|2Y|J|2Z|X|Y]]|9|@]|A|$]]|$1|1E|3|1F|5|6|7|30|8|@$I|31|J|32|X|Y]]|9|@]|A|$]]|$1|1G|3|-4|5|6|7|33|8|@]|9|@]|A|$]]]|1H|$1I|$5|1J|1K|1L|A|$1M|1N]]|1O|$5|1J|1K|1L|A|$1M|1P]]|1Q|$5|1J|1K|1L|A|$1M|1R]]|1S|$5|1T|1K|1U|A|$1V|1W|1X|-4]]]]

When you are loading the pickled model the source tree must match the one that used when the model was saved. So
<pre><code>ModuleNotFoundError: No module named 'models'
</code></pre>
expects this directory to be in your python path: <a href="https://github.com/WongKinYiu/yolov7/tree/main/models" rel="nofollow noreferrer">https://github.com/WongKinYiu/yolov7/tree/main/models</a>
To export to ONNX:
<ol>
<li>Clone the repo <a href="https://github.com/WongKinYiu/yolov7" rel="nofollow noreferrer">https://github.com/WongKinYiu/yolov7</a></li>
</ol>
<pre><code>git clone https://github.com/WongKinYiu/yolov7
</code></pre>
<ol start="2">
<li>Set the correct path to it.</li>
</ol>
<pre><code>import sys
sys.path.insert(0, './yolov7')
</code></pre>
or you can set <code>PYTHONPATH</code> environment variable
<ol start="3">
<li>Also you may need to have specific <code>torch</code> version. I've checked and it seems <code>torch==1.8.0</code> works fine</li>
</ol>
Example:
<pre class="lang-py prettyprint-override"><code>import torch
import sys

sys.path.insert(0, './yolov7')

device = torch.device('cpu')
model = torch.load('yolov7x.pt', map_location=device)['model'].float()
torch.onnx.export(model, torch.zeros((1, 3, 640, 640)), 'yolov7.onnx', opset_version=12)

</code></pre>
After that the model was exported to ONNX (visualized with <a href="https://github.com/lutzroeder/netron" rel="nofollow noreferrer">netron</a>):
<a href="https://i.stack.imgur.com/2kN4d.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/2kN4d.png" alt="enter image description here" /></a>
Usually it is better to save weights as <code>state_dict</code> and keep the source code that can reconstruct the <code>torch.nn.Module</code> so then you can safely use:
<code>model.load_state_dict(torch.load('weights.pt'))</code>

blocks|key|1358367|text|至于它的价值，我只能让它工作曾傑瑞操纵一些代码从另一个回购，加上答案，从@u1234x1234|type|unstyled|depth|inlineStyleRanges|entityRanges|data|1358368|import+torch
from+models.experimental+import+attempt_load
import+os

weights+='yolov7x.pt'
device+=+torch.device('cpu')

#+Load+model
model+=+attempt_load(weights,+map_location=device)+

torch.onnx.export(model,+torch.zeros((1,+3,+640,+640)),+'yolov7.onnx',+opset_version=12)|code-block|syntax|javascript|1358369|似乎与此函数attempt_load一起工作，这是另一个火把用户回购的一个文件。我不熟悉火把，不知道这是不是一个普通的火把文件.|offset|length|style|CODE|1358370|编辑|BOLD|1358371|查看open+:+Yolo7上的REPO：巨蟒|1358372|entityMap|0|LINK|mutability|MUTABLE|url|https://github.com/DoganK01/YOLOV7-OBJECT-COUNTER-V1.2/blob/main/models/experimental.py|1|https://github.com/OpenVINO-dev-contest/YOLOv7_OpenVINO_cpp-python^0|0|0|6|C|P|9|0|0|0|2|0|L|2|1|0^^$0|@$1|2|3|4|5|6|7|11|8|@]|9|@]|A|$]]|$1|B|3|C|5|D|7|12|8|@]|9|@]|A|$E|F]]|$1|G|3|H|5|6|7|13|8|@$I|14|J|15|K|L]]|9|@$I|16|J|17|1|18]]|A|$]]|$1|M|3|N|5|6|7|19|8|@$I|1A|J|1B|K|O]]|9|@]|A|$]]|$1|P|3|Q|5|6|7|1C|8|@]|9|@$I|1D|J|1E|1|1F]]|A|$]]|$1|R|3|-4|5|6|7|1G|8|@]|9|@]|A|$]]]|S|$T|$5|U|V|W|A|$X|Y]]|Z|$5|U|V|W|A|$X|10]]]]

For what its worth I could only get this to work jerry rigging some code from another repo, plus the answer from @u1234x1234
<pre><code>import torch
from models.experimental import attempt_load
import os

weights ='yolov7x.pt'
device = torch.device('cpu')

# Load model
model = attempt_load(weights, map_location=device) 

torch.onnx.export(model, torch.zeros((1, 3, 640, 640)), 'yolov7.onnx', opset_version=12)
</code></pre>
Appeared to work with this function <code>attempt_load</code> a file from <a href="https://github.com/DoganK01/YOLOV7-OBJECT-COUNTER-V1.2/blob/main/models/experimental.py" rel="nofollow noreferrer">this other pytorch user repo</a>. I am unfamiliar with pytorch and dont know if this is just a common pytorch file or not...
EDIT
See this REPO for Yolo7 on open VINO:
<a href="https://github.com/OpenVINO-dev-contest/YOLOv7_OpenVINO_cpp-python" rel="nofollow noreferrer">https://github.com/OpenVINO-dev-contest/YOLOv7_OpenVINO_cpp-python</a>

How to convert a pytorch model to ONNX? I am trying to use <a href="https://wandb.ai/wandb/common-ml-errors/reports/How-to-Save-and-Load-Models-in-PyTorch--VmlldzozMjg0MTE#load" rel="nofollow noreferrer">this method</a> on Python 3.7:
<pre><code>import torch

model = torch.load(&quot;./yolov7x.pt&quot;)

#torch.onnx.export(model, &quot;yolo_v7x.onnx&quot;)
</code></pre>
Even with the commented last line in the 3 lines of code, loading this errors out:
<pre><code>Traceback (most recent call last):
 File &quot;C:\Users\convert_onx.py&quot;, line 5, in &lt;module&gt;
 model = torch.load(&quot;./yolov7x.pt&quot;)
 File &quot;C:\Users\Python37\lib\site-packages\torch\serialization.py&quot;, line 594, in load
 return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
 File &quot;C:\Users\Python37\lib\site-packages\torch\serialization.py&quot;, line 853, in _load
 result = unpickler.load()
ModuleNotFoundError: No module named 'models'
</code></pre>
This is the <a href="https://github.com/WongKinYiu/yolov7#performance" rel="nofollow noreferrer">git repo</a> I am working with the Yolo Model 7x:
<a href="https://i.stack.imgur.com/F3ifo.png" rel="nofollow noreferrer"><img src="https://i.stack.imgur.com/F3ifo.png" alt="enter image description here" /></a>
Ultimate use case is to use this model on Intel's Open VINO toolkit that requires PyTorch models to be converted to ONYX.

convert pytorch model to ONNX

翻译质量差，导致语言生硬或混乱。

没有提供实际的解决方法或示例。

解答不清晰，无法理解或解决问题。

页面排版不美观，阅读体验差。

文章

问答

视频

学习中心

腾讯云实验室

直播

竞赛

腾讯云代码分析专区

腾讯iOA零信任安全管理系统专区

腾讯云架构师技术同盟交流圈

腾讯云数据库专区

腾讯云顾问专区

腾讯云原生专区

腾讯混元专区

腾讯云TCE专区

腾讯云Lighthouse专区

腾讯云HAI专区

腾讯云Edgeone专区

腾讯云存储专区

腾讯云智能专区

腾讯轻联专区 

腾讯云开发专区

TAPD专区

腾讯轻量云游戏服专区

腾讯云最具价值专家

腾讯云架构师技术同盟

腾讯云创作之星

腾讯云开发者先锋 

腾讯云代码助手

CODING DevOps

Cloud Studio

SDK中心

API中心

命令行工具

如何将火把模型转换为ONNX？我试图在Python3.7上使用：import torchmodel = torch.load("./yolov7x.pt")#torch.onnx.export(model, "yolo_v7x.onnx")即使在3行代码中有注释的最后一行，也要加载此错误：Traceback (most...

问将pytorch模型转换为ONNX
EN

Stack Overflow用户

社区

活动

圈层

关于

腾讯云开发者

热门产品

热门推荐

更多推荐

问将pytorch模型转换为ONNXEN