forked from nudt_dsp/netrans
318 lines
24 KiB
HTML
318 lines
24 KiB
HTML
<!DOCTYPE html>
|
||
|
||
<html lang="zh" data-content_root="./">
|
||
<head>
|
||
<meta charset="utf-8" />
|
||
<meta name="viewport" content="width=device-width, initial-scale=1.0" /><meta name="viewport" content="width=device-width, initial-scale=1" />
|
||
|
||
<title>netrans_cli 使用 — netrans 0.1 文档</title>
|
||
<link rel="stylesheet" type="text/css" href="_static/pygments.css?v=5ecbeea2" />
|
||
<link rel="stylesheet" type="text/css" href="_static/basic.css?v=b08954a9" />
|
||
<link rel="stylesheet" type="text/css" href="_static/alabaster.css?v=27fed22d" />
|
||
<script src="_static/documentation_options.js?v=52efc512"></script>
|
||
<script src="_static/doctools.js?v=9bcbadda"></script>
|
||
<script src="_static/sphinx_highlight.js?v=dc90522c"></script>
|
||
<link rel="index" title="索引" href="genindex.html" />
|
||
<link rel="search" title="搜索" href="search.html" />
|
||
<link rel="next" title="netrans_py 使用" href="netrans_py.html" />
|
||
<link rel="prev" title="快速入门" href="quick_start_guide.html" />
|
||
|
||
<link rel="stylesheet" href="_static/custom.css" type="text/css" />
|
||
|
||
|
||
|
||
|
||
|
||
</head><body>
|
||
|
||
|
||
<div class="document">
|
||
<div class="documentwrapper">
|
||
<div class="bodywrapper">
|
||
|
||
|
||
<div class="body" role="main">
|
||
|
||
<section id="netrans-cli">
|
||
<h1>netrans_cli 使用<a class="headerlink" href="#netrans-cli" title="Link to this heading">¶</a></h1>
|
||
<p>netrans_cli 是 netrans 进行模型转换的命令行工具，使用 ntrans_cli 完成模型转换的步骤如下：</p>
|
||
<ol class="simple">
|
||
<li><p>导入模型</p></li>
|
||
<li><p>生成并修改前处理配置文件 *_inputmeta.yml</p></li>
|
||
<li><p>量化模型</p></li>
|
||
<li><p>导出模型</p></li>
|
||
</ol>
|
||
<section id="id1">
|
||
<h2>netrans_cli 脚本<a class="headerlink" href="#id1" title="Link to this heading">¶</a></h2>
|
||
<table border="1" class="docutils">
|
||
<thead>
|
||
<tr>
|
||
<th style="text-align: left;">脚本</th>
|
||
<th>功能</th>
|
||
<th>使用</th>
|
||
</tr>
|
||
</thead>
|
||
<tbody>
|
||
<tr>
|
||
<td style="text-align: left;">load.sh</td>
|
||
<td>模型导入功能，将模型转换成 Pnna 支持的格式</td>
|
||
<td>load.sh model_name</td>
|
||
</tr>
|
||
<tr>
|
||
<td style="text-align: left;">config.sh</td>
|
||
<td>预处理模版生成功能，生成预处理模版，根据模型进行对于的修改</td>
|
||
<td>config.sh model_name</td>
|
||
</tr>
|
||
<tr>
|
||
<td style="text-align: left;">quantize.sh</td>
|
||
<td>量化功能， 对模型进行量化生成量化参数文件</td>
|
||
<td>quantize.sh model_name quantize_data_type</td>
|
||
</tr>
|
||
<tr>
|
||
<td style="text-align: left;">export.sh</td>
|
||
<td>导出功能，将量化好的模型导出成 Pnna 上可以运行的runtime</td>
|
||
<td>export.sh model_name quantize_data_type</td>
|
||
</tr>
|
||
</tbody>
|
||
</table><p><font color="#dd0000">对于不同框架下训练的模型，需要准备不同的数据，所有的数据都需要与模型放在同一个文件夹下，模型文件名和文件夹名需要保持一致。</font></p>
|
||
</section>
|
||
<section id="load-sh">
|
||
<h2>load.sh 模型导入<a class="headerlink" href="#load-sh" title="Link to this heading">¶</a></h2>
|
||
<p>使用 load.sh 导入模型</p>
|
||
<ul>
|
||
<li><p>用法： load.sh 以模型文件名命名的模型数据文件夹，例如：</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>load.sh<span class="w"> </span>lenet
|
||
</pre></div>
|
||
</div>
|
||
<p>"lenet"是文件夹名，也作为模型名和权重文件名。导入会打印相关日志信息，成功后会打印SUCESS。导入后lenet文件夹应该有"lenet.json"和"lenet.data"文件:</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>$<span class="w"> </span>ls<span class="w"> </span>-lrt<span class="w"> </span>lenet
|
||
total<span class="w"> </span><span class="m">3396</span>
|
||
-rwxr-xr-x<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">1727201</span><span class="w"> </span>Nov<span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="m">2018</span><span class="w"> </span>lenet.pb
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">553</span><span class="w"> </span>Nov<span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="m">2018</span><span class="w"> </span><span class="m">0</span>.jpg
|
||
-rwxr--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">6</span><span class="w"> </span>Apr<span class="w"> </span><span class="m">21</span><span class="w"> </span><span class="m">17</span>:04<span class="w"> </span>dataset.txt
|
||
-rw-rw-r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">69</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:19<span class="w"> </span>inputs_outputs.txt
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">5553</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:21<span class="w"> </span>lenet.json
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">1725178</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:21<span class="w"> </span>lenet.data
|
||
</pre></div>
|
||
</div>
|
||
</li>
|
||
</ul>
|
||
</section>
|
||
<section id="config-sh">
|
||
<h2>config.sh 预处理配置文件生成<a class="headerlink" href="#config-sh" title="Link to this heading">¶</a></h2>
|
||
<p>使用 config.sh 生成 inputmeta 文件</p>
|
||
<ul>
|
||
<li><p>config.sh 以模型文件名命名的模型数据文件夹，例如：</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>config.sh<span class="w"> </span>lenet
|
||
</pre></div>
|
||
</div>
|
||
<p>inputmeta 文件生成会打印相关日志信息，成功后会打印SUCESS。导入后lenet文件夹应该有 "lenet_inputmeta.yml" 文件:</p>
|
||
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span><span class="w"> </span>$<span class="w"> </span>ls<span class="w"> </span>-lrt<span class="w"> </span>lenet
|
||
total<span class="w"> </span><span class="m">3400</span>
|
||
-rwxr-xr-x<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">1727201</span><span class="w"> </span>Nov<span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="m">2018</span><span class="w"> </span>lenet.pb
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">553</span><span class="w"> </span>Nov<span class="w"> </span><span class="m">5</span><span class="w"> </span><span class="m">2018</span><span class="w"> </span><span class="m">0</span>.jpg
|
||
-rwxr--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">6</span><span class="w"> </span>Apr<span class="w"> </span><span class="m">21</span><span class="w"> </span><span class="m">17</span>:04<span class="w"> </span>dataset.txt
|
||
-rw-rw-r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">69</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:19<span class="w"> </span>inputs_outputs.txt
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">5553</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:21<span class="w"> </span>lenet.json
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">1725178</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:21<span class="w"> </span>lenet.data
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">948</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:35<span class="w"> </span>lenet_inputmeta.yml
|
||
</pre></div>
|
||
</div>
|
||
<p>可以看到，最终生成的是*.yml文件，该文件用于为Netrans中间模型配置输入层数据集合。<b>Netrans中的量化、推理、导出和图片转dat的操作都需要用到这个文件。因此，此步骤不可跳过。</b></p>
|
||
</li>
|
||
</ul>
|
||
<p>Inputmeta.yml文件结构如下：</p>
|
||
<div class="highlight-yaml notranslate"><div class="highlight"><pre><span></span><span class="nt">%YAML</span><span class="w"> </span><span class="m">1.2</span>
|
||
<span class="nn">---</span>
|
||
<span class="c1"># !!!This file disallow TABs!!!</span>
|
||
<span class="c1"># "category" allowed values: "image, undefined"</span>
|
||
<span class="c1"># "database" allowed types: "H5FS, SQLITE, TEXT, LMDB, NPY, GENERATOR"</span>
|
||
<span class="c1"># "tensor_name" only support in H5FS database</span>
|
||
<span class="c1"># "preproc_type" allowed types:"IMAGE_RGB, IMAGE_RGB888_PLANAR, IMAGE_RGB888_PLANAR_SEP, </span>
|
||
<span class="l l-Scalar l-Scalar-Plain">IMAGE_I420,</span><span class="w"> </span>
|
||
<span class="l l-Scalar l-Scalar-Plain"># IMAGE_NV12, IMAGE_YUV444, IMAGE_GRAY, IMAGE_BGRA, TENSOR"</span>
|
||
<span class="l l-Scalar l-Scalar-Plain">input_meta</span><span class="p p-Indicator">:</span>
|
||
<span class="w"> </span><span class="nt">databases</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">path</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">dataset.txt</span>
|
||
<span class="w"> </span><span class="nt">type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">TEXT</span>
|
||
<span class="w"> </span><span class="nt">ports</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">lid</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">data_0</span>
|
||
<span class="w"> </span><span class="nt">category</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">image</span>
|
||
<span class="w"> </span><span class="nt">dtype</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">float32</span>
|
||
<span class="w"> </span><span class="nt">sparse</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">false</span>
|
||
<span class="w"> </span><span class="nt">tensor_name</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="nt">layout</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">nhwc</span>
|
||
<span class="w"> </span><span class="nt">shape</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">50</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">224</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">224</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">3</span>
|
||
<span class="w"> </span><span class="nt">preprocess</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="nt">reverse_channel</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">false</span>
|
||
<span class="w"> </span><span class="nt">mean</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">103.94</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">116.78</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">123.67</span>
|
||
<span class="w"> </span><span class="nt">scale</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">0.017</span>
|
||
<span class="w"> </span><span class="nt">preproc_node_params</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="nt">preproc_type</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">IMAGE_RGB</span>
|
||
<span class="w"> </span><span class="nt">add_preproc_node</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">false</span>
|
||
<span class="w"> </span><span class="nt">preproc_perm</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">0</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">1</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">2</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">3</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="nt">lid</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">label_0</span>
|
||
<span class="w"> </span><span class="nt">redirect_to_output</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">true</span>
|
||
<span class="w"> </span><span class="nt">category</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">undefined</span>
|
||
<span class="w"> </span><span class="nt">tensor_name</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="nt">dtype</span><span class="p">:</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">float32</span>
|
||
<span class="w"> </span><span class="nt">shape</span><span class="p">:</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">1</span>
|
||
<span class="w"> </span><span class="p p-Indicator">-</span><span class="w"> </span><span class="l l-Scalar l-Scalar-Plain">1</span>
|
||
</pre></div>
|
||
</div>
|
||
<p>上面示例文件的各个参数解释：</p>
|
||
<div class="highlight-{table} notranslate"><div class="highlight"><pre><span></span>:widths: 20, 80
|
||
:align: left
|
||
| 参数 | 说明 |
|
||
| :--- | --- |
|
||
| input_meta | 预处理参数配置申明。 |
|
||
| databases | 数据配置，包括设置 path、type 和 ports 。|
|
||
| path | 数据集文件的相对（执行目录）或绝对路径。默认为 dataset.txt, 不建议修改。 |
|
||
| type | 数据集文件格式，固定为TEXT。 |
|
||
| ports | 指向网络中的输入或重定向的输入，目前只支持一个输入，如果网络存在多个输入，请与@ccyh联系。 |
|
||
| lid | 输入层的lid |
|
||
| category | 输入的类别。将此参数设置为以下值之一：image（图像输入）或 undefined（其他类型的输入）。 |
|
||
| dtype | 输入张量的数据类型，用于将数据发送到 Pnna 网络的输入端口。支持的数据类型包括 float32 和 quantized。 |
|
||
| sparse | 指定网络张量是否以稀疏格式存在。将此参数设置为以下值之一：true（稀疏格式）或 false（压缩格式）。 |
|
||
| tensor_name | 留空此参数 |
|
||
| layout | 输入张量的格式，使用 nchw 用于 Caffe、Darknet、ONNX 和 PyTorch 模型。使用 nhwc 用于 TensorFlow、TensorFlow Lite 和 Keras 模型。 |
|
||
| shape | 此张量的形状。第一维，shape[0]，表示每批的输入数量，允许在一次推理操作之前将多个输入发送到网络。如果batch维度设置为0，则需要从命令行指定--batch-size。如果 batch维度设置为大于1的值，则直接使用inputmeta.yml中的batch size并忽略命令行中的--batch-size。 |
|
||
| fitting | 保留字段 |
|
||
| preprocess | 预处理步骤和顺序。预处理支持下面的四个键，键的顺序代表预处理的顺序。您可以相应地调整顺序。 |
|
||
| reverse_channel | 指定是否保留通道顺序。将此参数设置为以下值之一：true（保留通道顺序）或 false（不保留通道顺序）。对于 TensorFlow 和 TensorFlow Lite 框架的模型使用 true。 |
|
||
| mean | 用于每个通道的均值。 |
|
||
| scale | 张量的缩放值。均值和缩放值用于根据公式 (inputTensor - mean) × scale 归一化输入张量。|
|
||
| preproc_node_params | 预处理节点参数，在 OVxlib C 项目案例中启用预处理任务 |
|
||
| add_preproc_node | 用于处理 OVxlib C 项目案例中预处理节点的插入。[true, false] 中的布尔值，表示通过配置以下参数将预处理层添加到导出的应用程序中。此参数仅在 add_preproc_node 参数设置为 true 时有效。|
|
||
| preproc_type | 预处理节点输入类型。 [IMAGE_RGB, IMAGE_RGB888_PLANAR,IMAGE_YUV420, IMAGE_GRAY, IMAGE_BGRA, TENSOR] 中的字符串值 |
|
||
| preproc_perm | 预处理节点输入的置换参数。 |
|
||
| redirect_to_output | 将database张量重定向到图形输出的特殊属性。如果为该属性设置了一个port，网络构建器将自动为该port生成一个输出层，以便后处理文件可以直接处理来自database的张量。 如果使用网络进行分类，则上例中的lid“input_0”表示输入数据集的标签lid。 您可以设置其他名称来表示标签的lid。 请注意，redirect_to_output 必须设置为 true，以便后处理文件可以直接处理来自database的张量。 标签的lid必须与后处理文件中定义的 labels_tensor 的lid相同。 [true, false] 中的布尔值。 指定是否将由张量表示的输入端口的数据直接发送到网络输出。true（直接发送到网络输出）或 false（不直接发送到网络输出）|
|
||
</pre></div>
|
||
</div>
|
||
<p>可以根据实际情况对生成的inputmeta文件进行修改。</p>
|
||
</section>
|
||
<section id="quantize-sh">
|
||
<h2>quantize.sh 模型量化<a class="headerlink" href="#quantize-sh" title="Link to this heading">¶</a></h2>
|
||
<p>如果我们训练好的模型的数据类型是float32的，为了使模型以更高的效率在Pnna上运行，我们可以对模型进行量化操作，量化操作可能会带来一定程度的精度损失。</p>
|
||
<ul class="simple">
|
||
<li><p>在netrans_cli目录下使用quantize.sh脚本进行量化操作。</p></li>
|
||
</ul>
|
||
<p>用法：./quantize.sh 以模型文件名命名的模型数据文件夹 量化类型，例如：</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>quantize.sh<span class="w"> </span>lenet<span class="w"> </span>uint8
|
||
</pre></div>
|
||
</div>
|
||
<p>支持的量化类型有：uint8、int8、int16</p>
|
||
</section>
|
||
<section id="export-sh">
|
||
<h2>export.sh 模型导出<a class="headerlink" href="#export-sh" title="Link to this heading">¶</a></h2>
|
||
<p>使用 export.sh 导出模型生成nbg文件。</p>
|
||
<p>用法：export.sh 以模型文件名命名的模型数据文件夹 数据类型，例如：</p>
|
||
<div class="highlight-bash notranslate"><div class="highlight"><pre><span></span>export.sh<span class="w"> </span>lenet<span class="w"> </span>uint8
|
||
</pre></div>
|
||
</div>
|
||
<p>导出支持的数据类型：float、uint8、int8、int16，其中使用uint8、int8、int16导出时需要先进行模型量化。导出的工程会在模型所在的目录下面的wksp目录里。
|
||
network_binary.nb文件在"asymmetric_affine"文件夹中：</p>
|
||
<div class="highlight-shell notranslate"><div class="highlight"><pre><span></span>ls<span class="w"> </span>-lrt<span class="w"> </span>lenet/wksp/asymmetric_affine/
|
||
-rw-r--r--<span class="w"> </span><span class="m">1</span><span class="w"> </span>hope<span class="w"> </span>hope<span class="w"> </span><span class="m">694912</span><span class="w"> </span>Jun<span class="w"> </span><span class="m">7</span><span class="w"> </span><span class="m">09</span>:55<span class="w"> </span>network_binary.nb
|
||
</pre></div>
|
||
</div>
|
||
<p>目前支持将生成的network_binary.nb文件部署到Pnna硬件平台。具体部署方法请参阅模型部署相关文档。</p>
|
||
</section>
|
||
<section id="id2">
|
||
<h2>使用示例<a class="headerlink" href="#id2" title="Link to this heading">¶</a></h2>
|
||
<p>请参照examples，examples 提供 <a class="reference external" href="./examples/caffe_model">caffe 模型转换示例</a>,<a class="reference external" href="./examples/darknet_model">darknet 模型转换示例</a>,<a class="reference external" href="./examples/tensorflow_model">tensorflow 模型转换示例</a>,<a class="reference external" href="./examples/onnx_model">onnx 模型转换示例</a>。</p>
|
||
</section>
|
||
</section>
|
||
|
||
|
||
</div>
|
||
|
||
</div>
|
||
</div>
|
||
<div class="sphinxsidebar" role="navigation" aria-label="Main">
|
||
<div class="sphinxsidebarwrapper">
|
||
<h1 class="logo"><a href="index.html">netrans</a></h1>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
<search id="searchbox" style="display: none" role="search">
|
||
<div class="searchformwrapper">
|
||
<form class="search" action="search.html" method="get">
|
||
<input type="text" name="q" aria-labelledby="searchlabel" autocomplete="off" autocorrect="off" autocapitalize="off" spellcheck="false" placeholder="Search"/>
|
||
<input type="submit" value="提交" />
|
||
</form>
|
||
</div>
|
||
</search>
|
||
<script>document.getElementById('searchbox').style.display = "block"</script><h3>导航</h3>
|
||
<p class="caption" role="heading"><span class="caption-text">Contents:</span></p>
|
||
<ul class="current">
|
||
<li class="toctree-l1"><a class="reference internal" href="quick_start_guide.html">快速入门</a></li>
|
||
<li class="toctree-l1 current"><a class="current reference internal" href="#">netrans_cli 使用</a><ul>
|
||
<li class="toctree-l2"><a class="reference internal" href="#id1">netrans_cli 脚本</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="#load-sh">load.sh 模型导入</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="#config-sh">config.sh 预处理配置文件生成</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="#quantize-sh">quantize.sh 模型量化</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="#export-sh">export.sh 模型导出</a></li>
|
||
<li class="toctree-l2"><a class="reference internal" href="#id2">使用示例</a></li>
|
||
</ul>
|
||
</li>
|
||
<li class="toctree-l1"><a class="reference internal" href="netrans_py.html">netrans_py 使用</a></li>
|
||
<li class="toctree-l1"><a class="reference internal" href="appendix.html">附录</a></li>
|
||
</ul>
|
||
|
||
<div class="relations">
|
||
<h3>Related Topics</h3>
|
||
<ul>
|
||
<li><a href="index.html">Documentation overview</a><ul>
|
||
<li>Previous: <a href="quick_start_guide.html" title="上一章">快速入门</a></li>
|
||
<li>Next: <a href="netrans_py.html" title="下一章">netrans_py 使用</a></li>
|
||
</ul></li>
|
||
</ul>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
|
||
</div>
|
||
</div>
|
||
<div class="clearer"></div>
|
||
</div>
|
||
<div class="footer">
|
||
©2025, ccyh.
|
||
|
||
|
|
||
Powered by <a href="https://www.sphinx-doc.org/">Sphinx 8.2.3</a>
|
||
& <a href="https://alabaster.readthedocs.io">Alabaster 1.0.0</a>
|
||
|
||
|
|
||
<a href="_sources/netrans_cli.md.txt"
|
||
rel="nofollow">Page source</a>
|
||
</div>
|
||
|
||
|
||
|
||
|
||
</body>
|
||
</html> |