<?xml version="1.0" encoding="UTF-8"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Yestino - The Signal · MPS Reduce</title><link>https://yestino.com/entities/mps-reduce-827ff0</link><description>Every event involving MPS Reduce</description><language>en</language><atom:link href="https://yestino.com/entities/mps-reduce-827ff0/feed.xml" rel="self" type="application/rss+xml"/><item><title>trunk/9d75d51a07ee04d24dde44688281b35ead889dce: [MPS] Reduce strided inputs in place for full reductions (#198646)</title><link>https://yestino.com/events/trunk-9d75d51a07ee04d24dde44688281b35ead889dce-mps-reduce-st-0156db</link><guid isPermaLink="true">https://yestino.com/events/trunk-9d75d51a07ee04d24dde44688281b35ead889dce-mps-reduce-st-0156db</guid><pubDate>Sat, 03 Oct 2026 03:41:06 GMT</pubDate><description>Since #198494 dropped the Strided pass-1 kernel, a full reduction ( dim=None ) over a non-contiguous input materialises a contiguous copy and then runs Flat ov…
Sources: PyTorch Releases</description></item></channel></rss>