<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title><![CDATA[point pillar &quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot;]]></title><description><![CDATA[<p><a href="https://github.com/k0suke-murakami/kitti_pretrained_point_pillars" target="_blank" rel="noopener noreferrer nofollow">https://github.com/k0suke-murakami/kitti_pretrained_point_pillars</a><br />
提供了初步训练的weight模型文件，想问下，这个怎么才能得到&quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot;，训练好point pillar之后得到的实际是 voxelnet-363855.tckpt 以及 adam_optimizer-363855.tckpt 这两个文件。</p>
]]></description><link>http://t.manaai.cn/topic/111/point-pillar-pre-onnx-as-well-as-rpn-onnx</link><generator>RSS for Node</generator><lastBuildDate>Thu, 15 Oct 2026 20:54:56 GMT</lastBuildDate><atom:link href="http://t.manaai.cn/topic/111.rss" rel="self" type="application/rss+xml"/><pubDate>Mon, 06 May 2019 01:20:33 GMT</pubDate><ttl>60</ttl><item><title><![CDATA[Reply to point pillar &quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot; on Mon, 06 May 2019 01:20:33 GMT]]></title><description><![CDATA[<p><a href="https://github.com/k0suke-murakami/kitti_pretrained_point_pillars" target="_blank" rel="noopener noreferrer nofollow">https://github.com/k0suke-murakami/kitti_pretrained_point_pillars</a><br />
提供了初步训练的weight模型文件，想问下，这个怎么才能得到&quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot;，训练好point pillar之后得到的实际是 voxelnet-363855.tckpt 以及 adam_optimizer-363855.tckpt 这两个文件。</p>
]]></description><link>http://t.manaai.cn/post/254</link><guid isPermaLink="true">http://t.manaai.cn/post/254</guid><dc:creator><![CDATA[Jean]]></dc:creator><pubDate>Mon, 06 May 2019 01:20:33 GMT</pubDate></item><item><title><![CDATA[Reply to point pillar &quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot; on Tue, 07 May 2019 02:26:37 GMT]]></title><description><![CDATA[<p>直接用不行吗？这个pointpillar不一定是second那个版本</p>
]]></description><link>http://t.manaai.cn/post/257</link><guid isPermaLink="true">http://t.manaai.cn/post/257</guid><dc:creator><![CDATA[刘看山]]></dc:creator><pubDate>Tue, 07 May 2019 02:26:37 GMT</pubDate></item><item><title><![CDATA[Reply to point pillar &quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot; on Wed, 08 May 2019 09:30:58 GMT]]></title><description><![CDATA[<p>@金天 他直接用从c++实现了前向推理，所以想问下怎么获得这两个文件</p>
]]></description><link>http://t.manaai.cn/post/265</link><guid isPermaLink="true">http://t.manaai.cn/post/265</guid><dc:creator><![CDATA[Jean]]></dc:creator><pubDate>Wed, 08 May 2019 09:30:58 GMT</pubDate></item><item><title><![CDATA[Reply to point pillar &quot;pre.onnx&quot; as well as &quot;rpn.onnx&quot; on Wed, 08 May 2019 10:13:20 GMT]]></title><description><![CDATA[<p><a class="plugin-mentions-user plugin-mentions-a" href="http://t.manaai.cn/uid/97">@Jean</a> 可以尝试从SECOND训练的模型转到onnx，然后用autoware里面的cuda文件进行预测。但是，我感觉比较难转换，second那个实现我还没有深入看，感觉依赖蛮多。再者，不知道这个仓库的作者为啥要分为两个模型分开导入，这不是端到端</p>
]]></description><link>http://t.manaai.cn/post/268</link><guid isPermaLink="true">http://t.manaai.cn/post/268</guid><dc:creator><![CDATA[刘看山]]></dc:creator><pubDate>Wed, 08 May 2019 10:13:20 GMT</pubDate></item></channel></rss>