<?xml version="1.0" encoding="UTF-8"?>
<rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:media="http://search.yahoo.com/mrss/" version="2.0"><channel><title>Dont Blame me.</title><link>https://cocolat.cc/</link><atom:link href="https://cocolat.cc/rss.xml" rel="self" type="application/rss+xml"/><description>Dont Blame me.</description><generator>Halo v2.25.4</generator><language>zh-cn</language><lastBuildDate>Wed, 30 Sep 2026 10:05:41 GMT</lastBuildDate><item><title><![CDATA[从检索策略到论文研读：用 Codex 与 Undermind 搭建科研文献工作流（上）]]></title><link>https://cocolat.cc/archives/codex-undermind-zotero-research-workflow-part-1</link><description><![CDATA[<img src="https://cocolat.cc/plugins/feed/assets/telemetry.gif?title=%E4%BB%8E%E6%A3%80%E7%B4%A2%E7%AD%96%E7%95%A5%E5%88%B0%E8%AE%BA%E6%96%87%E7%A0%94%E8%AF%BB%EF%BC%9A%E7%94%A8%20Codex%20%E4%B8%8E%20Undermind%20%E6%90%AD%E5%BB%BA%E7%A7%91%E7%A0%94%E6%96%87%E7%8C%AE%E5%B7%A5%E4%BD%9C%E6%B5%81%EF%BC%88%E4%B8%8A%EF%BC%89&amp;url=/archives/codex-undermind-zotero-research-workflow-part-1" width="1" height="1" alt="" style="opacity:0;">从经过脱敏的遥感研究场景出发，以 Codex 制定策略、Undermind 检索与核验、Zotero 整理结果，跑通最小科研文献工作流。]]></description><guid isPermaLink="false">/archives/codex-undermind-zotero-research-workflow-part-1</guid><dc:creator>Administrator</dc:creator><category>AI与GIS</category><pubDate>Fri, 14 Aug 2026 11:31:37 GMT</pubDate></item><item><title><![CDATA[从 RGB 到 13 波段：用 TorchGeo 和 EuroSAT 理解多光谱迁移学习]]></title><link>https://cocolat.cc/archives/day6-torchgeo-eurosat-multispectral-transfer-learning</link><description><![CDATA[<img src="https://cocolat.cc/plugins/feed/assets/telemetry.gif?title=%E4%BB%8E%20RGB%20%E5%88%B0%2013%20%E6%B3%A2%E6%AE%B5%EF%BC%9A%E7%94%A8%20TorchGeo%20%E5%92%8C%20EuroSAT%20%E7%90%86%E8%A7%A3%E5%A4%9A%E5%85%89%E8%B0%B1%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0&amp;url=/archives/day6-torchgeo-eurosat-multispectral-transfer-learning" width="1" height="1" alt="" style="opacity:0;">使用 EuroSAT100 对比 RGB + ImageNet 与 13 波段 + Sentinel-2 预训练两条迁移学习路线，并解释 TorchGeo 数据接口、波段顺序、通道匹配、logits、Softmax 和小样本验证结果。]]></description><guid isPermaLink="false">/archives/day6-torchgeo-eurosat-multispectral-transfer-learning</guid><dc:creator>Administrator</dc:creator><category>GeoAI</category><pubDate>Thu, 13 Aug 2026 04:12:08 GMT</pubDate></item><item><title><![CDATA[从一条样本到最佳模型：PyTorch 二分类训练机制]]></title><link>https://cocolat.cc/archives/pytorch-training-mechanics-beginner-guide</link><description><![CDATA[<img src="https://cocolat.cc/plugins/feed/assets/telemetry.gif?title=%E4%BB%8E%E4%B8%80%E6%9D%A1%E6%A0%B7%E6%9C%AC%E5%88%B0%E6%9C%80%E4%BD%B3%E6%A8%A1%E5%9E%8B%EF%BC%9APyTorch%20%E4%BA%8C%E5%88%86%E7%B1%BB%E8%AE%AD%E7%BB%83%E6%9C%BA%E5%88%B6&amp;url=/archives/pytorch-training-mechanics-beginner-guide" width="1" height="1" alt="" style="opacity:0;">使用二维平衡二分类数据，从 sample、batch 和 logits 开始，完整拆解 PyTorch 的训练、验证、checkpoint 保存与恢复流程。]]></description><guid isPermaLink="false">/archives/pytorch-training-mechanics-beginner-guide</guid><dc:creator>Administrator</dc:creator><category>深度学习基础</category><pubDate>Wed, 12 Aug 2026 02:27:30 GMT</pubDate></item><item><title><![CDATA[从 GIS 栅格到 PyTorch 张量：我的 GeoAI 学习笔记（Day 1—Day 3）]]></title><link>https://cocolat.cc/archives/geoai-day1-day3-notes</link><description><![CDATA[<img src="https://cocolat.cc/plugins/feed/assets/telemetry.gif?title=%E4%BB%8E%20GIS%20%E6%A0%85%E6%A0%BC%E5%88%B0%20PyTorch%20%E5%BC%A0%E9%87%8F%EF%BC%9A%E6%88%91%E7%9A%84%20GeoAI%20%E5%AD%A6%E4%B9%A0%E7%AC%94%E8%AE%B0%EF%BC%88Day%201%E2%80%94Day%203%EF%BC%89&amp;url=/archives/geoai-day1-day3-notes" width="1" height="1" alt="" style="opacity:0;">用三天时间打通 GIS 数据体检、多光谱遥感分析，以及栅格影像到 NumPy 数组和 PyTorch 张量的基础流程。]]></description><guid isPermaLink="false">/archives/geoai-day1-day3-notes</guid><dc:creator>Administrator</dc:creator><category>GeoAI 学习</category><pubDate>Wed, 12 Aug 2026 02:24:48 GMT</pubDate></item><item><title><![CDATA[预训练 ResNet18 迁移学习：从数据加载到最佳 Checkpoint 的完整实验记录]]></title><link>https://cocolat.cc/archives/day5-resnet18-transfer-learning</link><description><![CDATA[<img src="https://cocolat.cc/plugins/feed/assets/telemetry.gif?title=%E9%A2%84%E8%AE%AD%E7%BB%83%20ResNet18%20%E8%BF%81%E7%A7%BB%E5%AD%A6%E4%B9%A0%EF%BC%9A%E4%BB%8E%E6%95%B0%E6%8D%AE%E5%8A%A0%E8%BD%BD%E5%88%B0%E6%9C%80%E4%BD%B3%20Checkpoint%20%E7%9A%84%E5%AE%8C%E6%95%B4%E5%AE%9E%E9%AA%8C%E8%AE%B0%E5%BD%95&amp;url=/archives/day5-resnet18-transfer-learning" width="1" height="1" alt="" style="opacity:0;">使用 PyTorch 与预训练 ResNet18 完成蚂蚁和蜜蜂图像分类的完整学习实验记录。]]></description><guid isPermaLink="false">/archives/day5-resnet18-transfer-learning</guid><dc:creator>Administrator</dc:creator><category>学习日志</category><pubDate>Mon, 10 Aug 2026 03:33:44 GMT</pubDate></item></channel></rss>