Xcode: Delete row shortcut

Edit the plist file

It's at /Applications/Xcode.app/Contents/Frameworks/IDEKit.framework/Resources/IDETextKeyBindingSet.plist

Note that:

  • the linked answer suggests using XCode for this, but I find that this crashes XCode
  • this path is for Lion - it may vary for other OS versions
  • you may have to sudo or change permissions to edit this file
  • close XCode before editing

Add a new <dict> element

Add this text just ABOVE the close of the <dict> at the bottom of this file

    <key>Custom</key>
    <dict>
      <key>Delete Current Line In One Hit</key>
      <string>moveToEndOfLine:, deleteToBeginningOfLine:, deleteToEndOfParagraph:</string> 
    </dict>

Note that:

  • This is inside the existing <dict> so there is a </dict></plist> after this
  • This associates the named action "Delete Current Line In One Hit" to the three key actions in the string. I played around with different options here to get the right combination. For instance, if you try moveToBeginningOfLine, deleteToEndOfLine instead of vice versa, then using it on empty lines will delete the line and the entire next line. Which is unpleasant

Save this and open XCode

Add a new key binding in XCode

See the screenshots for this in the linked answer or do this:

  1. Open XCode Preferences
  2. Select Key Bindings
  3. Click on All
  4. In the SEarch box type "Delete" and search for "Delete Current Line in One Hit"
  5. Add your new binding.

This should do it but for one caveat. I could never get XCode to save the new binding. The same problem is described here though my workaround had a slight variation:

  1. Create a duplicate of the Default bindings (hit the + button at the bottom left of the Bindings window and create a duplicate)
  2. Set the new binding in the duplicate
  3. Close the preference dialog (that's the variation!)
  4. Open preferences again and make a new duplicate of the old duplicate and delete the old duplicate.

What a rigmarole, eh? But it works. Now I have Ctrl+D mapped to delete the entire current line in XCode 4.3 on Lion.


也可以参照 http://www.cocoachina.com/industry/20140613/8816.html

资源下载链接为: https://pan.quark.cn/s/140386800631 通用大模型文本分类实践的基本原理是,借助大模型自身较强的理解和推理能力,在使用时需在prompt中明确分类任务目标,并详细解释每个类目概念,尤其要突出类目间的差别。 结合in-context learning思想,有效的prompt应包含分类任务介绍及细节、类目概念解释、每个类目对应的例子和待分类文本。但实际应用中,类目和样本较多易导致prompt过长,影响大模型推理效果,因此可先通过向量检索缩小范围,再由大模型做最终决策。 具体方案为:离线时提前配置好每个类目的概念及对应样本;在线时先对给定query进行向量召回,再将召回结果交给大模型决策。 该方法不更新任何模型参数,直接使用开源模型参数。其架构参考GPT-RE并结合相关实践改写,加入上下文学习以提高准确度,还使用BGE作为向量模型,K-BERT提取文本关键词,拼接召回的相似例子作为上下文输入大模型。 代码实现上,大模型用Qwen2-7B-Instruct,Embedding采用bge-base-zh-v1.5,向量库选择milvus。分类主函数的作用是在向量库中召回相似案例,拼接prompt后输入大模型。 结果方面,使用ICL时accuracy达0.94,比bert文本分类的0.98低0.04,错误类别6个,处理时添加“家居”类别,影响不大;不使用ICL时accuracy为0.88,错误58项,可能与未修改prompt有关。 优点是无需训练即可有较好结果,例子优质、类目界限清晰时效果更佳,适合围绕通用大模型api打造工具;缺点是上限不高,仅针对一个分类任务部署大模型不划算,推理速度慢,icl的token使用多,用收费api会有额外开销。 后续可优化的点是利用key-bert提取的关键词,因为核心词语有时比语意更重要。 参考资料包括
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值