<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Ossdd on DeepLearning.Earth</title><link>https://deeplearning.earth/tags/ossdd/</link><description>Recent content in Ossdd on DeepLearning.Earth</description><generator>Hugo -- gohugo.io</generator><language>en</language><lastBuildDate>Thu, 15 Oct 2026 06:00:00 +0700</lastBuildDate><atom:link href="https://deeplearning.earth/tags/ossdd/index.xml" rel="self" type="application/rss+xml"/><item><title>OSSDD — Sentinel-1 ships, finetuned with RTMDet, Faster R-CNN, and FCOS</title><link>https://deeplearning.earth/posts/2026-10-15_ossdd_sentinel1_ships_rtmdet_faster_rcnn_fcos/</link><pubDate>Thu, 15 Oct 2026 06:00:00 +0700</pubDate><guid>https://deeplearning.earth/posts/2026-10-15_ossdd_sentinel1_ships_rtmdet_faster_rcnn_fcos/</guid><description>We took a detector that had learned ships in daylight, and taught it Sentinel-1 radar.
Twelve epochs of RTMDet-S on the VH chips of OSSDD reached 88.54% AP50 on the held-out test. It found 95.5% of the ships, in 45 minutes.
Five ships in a harbor, bright land around the water. RTMDet-S at score 0.3 draws five boxes and matches all five. The same recipe with the other two heads is below.</description></item></channel></rss>