EU4-42: Attending a networking event

最新推荐文章于 2025-03-02 18:46:31 发布
原创 最新推荐文章于 2025-03-02 18:46:31 发布 · 819 阅读 · 0 ·
CC 4.0 BY-SA版权
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。

本文通过一场偶遇的对话展示了如何在初次见面时运用合适的对话主题、闲聊技巧以及找到共同话题的方法，强调了避免敏感话题的重要性。

Vocabulary

[NINA] Haven't we met somewhere before?
[WOMAN] No, I don't think so.
[NINA] Oh. Sorry.
[NINA] So, are you from around here?
[WOMAN] Oh, not really. My husband and I moved here a few weeks ago.
[NINA]Oh, really? So did I，I think the area's lovely.
[WOMAN] Oh, me, too.
[WOMAN] We're going to open a book shop down the road in about a month.
[WOMAN] You should come to our opening.
[NINA] I will. Thanks.
[NINA] So what did you do before?
[WOMAN] Oh, well, I worked at Crow and Dunny's. Do you know it?
[NINA] I used to work there!
[NINA] I knew you looked familiar.
[WOMAN] Oh, right.
[WOMAN] Did you know Richard Moat?
[NINA] Oh, yes, Isn't he the worst?
[NINA] I know everybody who worked for him hated him!
[WOMAN] That's my husband. We just got married.
[NINA] Oh! not that Richard Moat! There was a ...oh, what was his name? Richard... .
Key words
  • Haven’t we met somewhere before?
  • Are you form around here?
  • Do you know it?
  • I knew you looked familiar.
  • Did you know Richard Moat?

Grammar

1. Topics for conversation

对话的主题合适与否，并没有影响规定。这往往取决于文化，社会和个人观点。因此与不熟悉人交谈时还是谨慎为好。
努力避免谈及诸如政治、宗教、金钱、爱情容易引起冲突的话题。相反，应当选择有趣且不会冒犯对方的话题。

A: What kind of sports do you like?
B: I'm a big soccer fan.
A: Did you see the game last night?
B: Yes! It was really exciting, wasn't it?
A: Do you enjoy traveling?
B: Yes, I do. I went to India last year. I love it!
A: The weather's nice right now, isn't it?
B: Yes, it is. I love the summer here.
2. Making small talk

使用下面这些主题在聚会和社交活动上进行简短而礼貌的对话。这称作闲聊。

Wow - it’s so hot outside!
This event is very well-planned.
The food’s not too bad.
So, what do you do for a living?
I’m with Crump and Company. Where are you?
Are you here with family?
Are you taking any time off this summer?

3. Finding common ground

与刚在社交晚会上刚见到的某人建立共识是有好处的，如果你认为你以前见过对方但是不确定，可以使用下面表达方式。

We’ve met somewhere before, haven’t we?
You look familiar.

使用下述表达确实你们是否来自同一个地方。

Are you from around here?
Have you always lived in the area?

使用下面表达确认你们是否知道同一个人或者地方。

I was the manager at Zutz Bajery. Are you familiar with it?
Did you ever work with Robin Meyer?

【论文浅读】《MAP-Net: Multi Attending Path Neural Network for Building Footprint Extraction from Remote S》
卷子的博客
05-26 2538
创新思想： 文章提出了多路径神经网络(Multi Attending Path Neural Network, MAP-Net)，用于精确提取多尺度建筑足迹和精确边界。这个网络通过多条平行路径学习保留多尺度特征的空间位置，并且每个阶段都以固定频率提取高层语意特征。 主要原理： MAP-Net的整体结构由三个模块组成： (a)多径特征提取网络，在保留空间细节信息的同时提取多尺度高层语义特征； (b)基于注意的多尺度特征自适应压缩和全局空间池增强模块； ©上采样和构建足迹提取模块； Conv块：由一系列用于提
论文《SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks》
L_YXDE的博客
07-03 3267
论文地址：https://arxiv.org/pdf/2006.10503.pdf SE3-Transformers等变性摘要背景以及相关工作一、自注意力机制（The Attention Mechanism）排列等变性（Permutation equivariance）二、图神经网络（Graph Neural Networks）三、等变性（ Equivariance）Group RepresentationsTensor Field Networks方法3.1 Neighbourhoods3.2 The S
ICLR2025 Spatial-Mamba：基于结构感知状态融合的有效视觉状态空间模型
u014546828的博客
03-02 1452
随后，该特征图被输入到四个连续的阶段中。与图2(a)中的原始Mamba相比，我们可以看到原始状态变量 x_t​ 直接受其前一状态 x_{t−1​} 的影响，而结构感知状态变量ht​通过融合机制整合了额外的邻域状态变量x_{ρ_1(t)}, x_{ρ_2(t)}, . . . , x_{ρ_k(t)}​，其中K=∣Ω∣表示邻域集合的大小。其中，k^d_{i,j} 表示膨胀因子 d 在位置 (i,j) 的滤波器权重，x_{t+iw+j​} 表示位于位置 (i,j) 的状态 x_t​的邻居，w 表示图像的宽度。
NLP之GPT-3：《 Language Models are Few-Shot Learners》的翻译与解读
热门推荐
头部AI社区如有邀博主AI主题演讲请私信—心比天高，仗剑走天涯，保持热爱，奔赴向梦想！低调，专注，谦虚，自律，反思，成长，还算比较正能量的博主，公益免费传播…内心特别想在AI界做出一些可以推进历史进程影响力的技术(兴趣使然，有点小情怀，也有点使命感呀
07-29 1万+
​ NLP之GPT-3：《 Language Models are Few-Shot Learners》的翻译与解读 目录 相关文章 《GPT-3: Language Models are Few-Shot Learners》的翻译与解读 Abstract 摘要 1 Introduction 介绍 2 Approach方法 3 Results 结果 4 Measuring and Preventing Memorization Of Ben
ZCMU--4797: Palindromic Times(C语言)
小小小Why的博客
04-09 506
Tattah is asleep if and only if Tattah is attending a lecture.This is a well-known formula among Tattah's colleagues. On a Wednesday afternoon, Tattah was attending Professor HH's lecture. At12:21, right before falling asleep, he was staring at the digi..
R-VQA: Learning Visual Relation Facts with Semantic Attention for Visual Question Answering
Monster_ixx的博客
06-01 2004
摘要： 最近，视觉问答（VQA）已经成为多模式学习中最重要的任务之一，因为它需要理解视觉和文本模式。现有方法主要依靠提取图像和问题特征来通过多模态融合或注意机制来学习它们的联合特征嵌入。最近的一些研究利用外部VQA独立模型来检测图像中的候选实体或属性，其作为与VQA任务互补的语义知识。但是，这些候选实体或属性可能与VQA任务无关，并且语义容量有限。为了更好地利用图像中的语义知识，我们提出了一种新的...
A2N: Attending to Neighbors for Knowledge Graph Inference笔记
summer_rain18的博客
12-14 1285
作者提出了一种基于注意力机制的方法学习依赖于query的实体表示，自适应地组合实体相关的邻居。 多跳推理上的优势 1. Motivation: 同一个实体要预测的关系不同时，关注的邻居子集不同；以及隐含模型如何推理多跳（当预测的尾实体为US时，两跳推理places_lived，country of these places） 2. A2N model （1）随机初始化实体和关系的表示,, 给定实体，对应的邻居集合为, 每个邻居的表示通过连接实体和关系的初始表示并使用线性变换后得到，即 （2）对于当前
《AM-Bi-LSTM: Adaptive Multi-Modal Bi-LSTM for Sequential Recommendation》
buyaotutou的博客
07-29 690
文章链接
Fashion-MNIST: Get Start with Deep Learning in Keras (Tensorflow based)
Mango_Holi的博客
05-25 565
The code is a material forHoli's tutorial of 'Introduction to Machine Learning', Code and Coffee,GDG Suzhou. Thanks for Tensorflow team to provide the tutorial for us and then make it much easier...
ipython-sql:适用于IPython的%% sql魔术，有望发展成为完整SQL客户端
04-28
ipythonSQL 作者： 凯瑟琳·德芙琳（Catherine Devlin）， 引入了％sql（或%% sql）魔术。...Out [ 2 ]: [( u'Alice' , u'Alice' , u'ALICE' , u'a lady attending on Princess Katherine' , 22 )] In [ 3 ]: res
【KDD 2020】 HGCN: A Heterogeneous Graph Convolutional Network-Based Deep Learning Model Toward Collec
u011754184的博客
09-14 1888
背景挑战 Collective classification [9, 25] aiming to collectively categorize the inter-connected entities in networked data, plays an important role in the various in-depth analysis tasks [3, 23, 31], such as genre identification in a movieLens network, socia
EU4-44: Summarizing a networking event
豆包不黏
08-30 1231
Business socializing
EU4-43: Making a business contact
豆包不黏
08-29 990
Business socializing
EU4-12: Passing on information
豆包不黏
07-24 908
Information Present tense Asking for and giving advice
EU5-12:Talking about the future technology
豆包不黏
10-07 779
The future
EU4-53: Making a proposal
豆包不黏
09-02 598
Business proposals
EU4-14: Summarizing a meeting
豆包不黏
07-27 584
Reported, Closing a meeting, Anyone,Anything
EU4-51: Describing a business situation
豆包不黏
08-31 579
Business socializing
select '1', count(*) from hd_patient a left join zylc.emr_ma_medical_staff_info b on a.mr_id=b.mr_id where a.ward_admission_date>=to_date('2025-01-01 00:00:00','yyyy-mm-dd hh24:mi:ss') and a.ward_admission_date<=to_date('2025-11-14 00:00:00','yyyy-mm-dd hh24:mi:ss') and ( b.kzr_doctor_code ='1616' or b.zr_doctor_code ='1616' or b.zz_doctor_code='1616' or b.zy_doctor_code='1616') union all select '2', count(*) from hd_patient a left join zylc.emr_ma_medical_staff_info b on a.mr_id=b.mr_id where a.ward_admission_date>=to_date('2025-01-01 00:00:00','yyyy-mm-dd hh24:mi:ss') and a.ward_admission_date<=to_date('2025-11-14 00:00:00','yyyy-mm-dd hh24:mi:ss') and ( b.kzr_doctor_code ='1616' or b.zr_doctor_code ='1616' or b.zz_doctor_code='1616' or b.zy_doctor_code='1616')
最新发布
11-18
你提供的 SQL 语句如下： ```sql SELECT '1', COUNT(*) FROM hd_patient a LEFT JOIN zylc.emr_ma_medical_staff_info b ON a.mr_id = b.mr_id WHERE a.ward_admission_date >= TO_DATE('2025-01-01 00:00:00', 'YYYY-MM-DD HH24:MI:SS') AND a.ward_admission_date <= TO_DATE('2025-11-14 00:00:00', 'YYYY-MM-DD HH24:MI:SS') AND (b.kzr_doctor_code = '1616' OR b.zr_doctor_code = '1616' OR b.zz_doctor_code = '1616' OR b.zy_doctor_code = '1616') UNION ALL SELECT '2', COUNT(*) FROM hd_patient a LEFT JOIN zylc.emr_ma_medical_staff_info b ON a.mr_id = b.mr_id WHERE a.ward_admission_date >= TO_DATE('2025-01-01 00:00:00', 'YYYY-MM-DD HH24:MI:SS') AND a.ward_admission_date <= TO_DATE('2025-11-14 00:00:00', 'YYYY-MM-DD HH24:MI:SS') AND (b.kzr_doctor_code = '1616' OR b.zr_doctor_code = '1616' OR b.zz_doctor_code = '1616' OR b.zy_doctor_code = '1616'); ``` --- ## ❌ 问题分析 虽然这次你在两个 `SELECT` 中加了 `'1'` 和 `'2'` 作为标签，但 **两个子查询完全相同**，所以结果会是： | label | count(*) | |-------|----------| | 1 | N | | 2 | N | 即：同一个数字被显示两次，只是前缀不同。这通常不是业务需要的逻辑 —— 很可能是误操作或测试残留。 --- ## ✅ 正确理解与建议 ### 🔍 可能的意图推测： 你可能是想对比两种情况（比如不同医生、不同科室、不同时间段、不同患者类型），但现在两个查询一模一样，没有意义。 --- ## ✅ 推荐修正方案（假设真实需求是“统计某医生在指定时间内的收治患者数”） ### ✅ 正确写法（单次查询即可）： ```sql SELECT 'Total' AS query_label, COUNT(*) AS patient_count FROM hd_patient a INNER JOIN zylc.emr_ma_medical_staff_info b ON a.mr_id = b.mr_id WHERE a.ward_admission_date >= TO_DATE('2025-01-01', 'YYYY-MM-DD') AND a.ward_admission_date < TO_DATE('2025-11-15', 'YYYY-MM-DD') -- 更安全的时间范围 AND ( b.kzr_doctor_code = '1616' OR b.zr_doctor_code = '1616' OR b.zz_doctor_code = '1616' OR b.zy_doctor_code = '1616' ); ``` > 输出示例： > > | QUERY_LABEL | PATIENT_COUNT | > |-------------|---------------| > | Total | 342 | --- ## 🚫 为什么不应该用 `UNION ALL` 写两个一样的查询？ - 导致结果重复，容易误导前端展示或报表统计。 - 浪费数据库资源，执行两遍相同的复杂查询。 - 如果后续要做聚合处理（如求和、导出 Excel），需额外去重或识别标签。 --- ## ✅ 什么时候才该使用 `UNION ALL`？ 当你有 **多个不同条件的分组统计** 时，例如： ### 示例：分别统计主治医生和住院医生身份下的接诊人数 ```sql SELECT 'Attending_Doctor' AS role, COUNT(*) AS cnt FROM hd_patient a INNER JOIN zylc.emr_ma_medical_staff_info b ON a.mr_id = b.mr_id WHERE a.ward_admission_date >= DATE '2025-01-01' AND a.ward_admission_date < DATE '2025-11-15' AND b.zr_doctor_code = '1616' UNION ALL SELECT 'Resident_Doctor' AS role, COUNT(*) AS cnt FROM hd_patient a INNER JOIN zylc.emr_ma_medical_staff_info b ON a.mr_id = b.mr_id WHERE a.ward_admission_date >= DATE '2025-01-01' AND a.ward_admission_date < DATE '2025-11-15' AND b.zy_doctor_code = '1616'; ``` 输出： | ROLE | CNT | |--------------------|-----| | Attending_Doctor | 89 | | Resident_Doctor | 76 | 这才是 `UNION ALL` 的合理用途。 --- ## ⚠️ 其他优化建议 1. **替换 `LEFT JOIN` 为 `INNER JOIN`** 因为你依赖 `b` 表字段做筛选，若 `b` 为空则无法满足条件，因此实际效果等同于内连接，直接写 `INNER JOIN` 更清晰高效。 2. **日期写法简化（Oracle 支持）** ```sql a.ward_admission_date >= DATE '2025-01-01' AND a.ward_admission_date < DATE '2025-11-15' ``` 或使用 `TO_DATE` 带格式： ```sql TO_DATE('2025-11-15', 'YYYY-MM-DD') -- 默认时间为 00:00:00 ``` 3. **避免时间边界误差** 使用 `< 下一天 00:00:00` 比 `<= 当天 23:59:59` 更精确，防止毫秒/微秒数据丢失。 4. **添加去重逻辑（如需按病人去重）** 若一个病人多次入院，你想只算一次： ```sql SELECT COUNT(DISTINCT a.patient_id) FROM ... ``` --- ###
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符
 
红包 添加红包
表情包 插入表情
表情包 代码片
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
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
实付元
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