Aryna Sylenko

Take a look at this… JOIN: Used to combine rows from two or more tables based on a related column between them. There are different types of joins, such as INNER JOIN, LEFT JOIN, RIGHT JOIN, and FULL OUTER JOIN. UNION: Used to combine the result sets of two or more SELECT queries into a single result set. DISTINCT: Used to retrieve unique rows from a result set by eliminating duplicate IN: Used to specify multiple values in a WHERE clause. BETWEEN: Used to filter data within a specific range. LIKE: Used for pattern matching in a WHERE clause. COUNT: An aggregate function that returns the number of rows in a result set or the number of occurrences of a particular value. SUM, AVG, MIN, MAX: Other common aggregate functions used to perform calculations on data in a result set. #Dataanalytics #SQL #MySql #Growth

8

6 commentaren

Here's how I landed my Dream Data Science Job. Even without experience: It's never been easier to apply for jobs. You set-up your LinkedIn profile and spam the easy apply button. This seems great but a wise philosopher said 'Hard choices, easy life. Easy choices, hard life'. It's easy to one-click apply. But if you don't think about why you care about that company. If you don't research what the company does and their hopes, problems and achievements. If you don't show this in a cover letter or well-written outreach message. You make things harder for yourself in the future - by getting ghosted again, with no idea why you didn't get a first interview. LinkedIn easy apply is convenient, but it's not fruitful for getting interviews. Whilst researching a company and writing a good outreach message can take more time, it's actually quite easy when you know what you're doing. That's how I got my job at a leading Health Insurance company with 1million customers, even though I'd never worked in Insurance before. So stop easy-applying for jobs, and show people why you should get an interview.

9

☝️ Ready to level up as a data scientist? Join us in the #MakingAIFrictionless journey! 🚀 As part of our mission to help data scientists learn MLOps, we’ve teamed up with Zindi to launch an MLOps competition designed to elevate your skills. To help ease the process, we created a 3-part video series to get you from development to deployment 🚀 In part 1 of our 3-part series, we're happy to introduce you to the fundamentals of Docker. This video guide is your first step toward mastering model packaging and seeing your AI impact in real scenarios. Don’t just build models — deploy them with confidence with Melio AI! 🔗Link to video: https://lnkd.in/db6y8Hh6 Let's keep pushing the boundaries of AI together! ♻️ If you found this helpful, consider resharing and follow Melio AI for more content like this. P.S. What part of Docker basics are you most excited to learn about? Comment below! #MLOps #MachineLearning #Highwind #MelioAI #Zindi #TechInnovation #AI #ML #DataScience

27

1 commentaar

Excited to have earned my "Intermediate SQL" certification from DataCamp! This course enhanced my skills in error handling, debugging, writing readable code, selecting and querying data, filtering and summarizing data, and sorting and grouping data. Looking forward to applying these skills to drive data-driven insights and projects. #SQL #Datacamp #learning

19

3 commentaren

Data aggregation methods may come in as iteratively constant until you find cases where you have to customize them to match your case.

2

Ever wondered what makes a Data Scientist tick? 🧠 It's a unique blend of caffeine-fueled curiosity, hawk eyes for detail, and a passion for unraveling the mysteries hidden within data. 📊🔍 At Data Genomix, our Data Scientists are the heart and soul of our operation, transforming raw information into actionable insights that drive business growth. 📈🚀 #DataScientist #DataGenomix #DataDriven #MarketingAnalytics

1

Confidence Interval 1. What is a Confidence Interval? A confidence interval is a range of values that is likely to contain the true value of a population parameter with a certain degree of confidence. It is used to estimate population parameters based on sample data. 2. Components of a Confidence Interval Point estimate: A single value calculated from sample data that estimates the population parameter (e.g., sample mean for population mean). Margin of error: The amount added and subtracted from the point estimate to determine the interval's endpoints. Confidence level: The probability that the interval will contain the true population parameter. 3. Interpretation A 95% confidence interval means that if we were to repeat the sampling process many times, we would expect 95% of the intervals to contain the true population parameter. 4. Factors Affecting Confidence Interval Width Sample size: Larger sample sizes generally lead to narrower intervals. Confidence level: Higher confidence levels (e.g., 99%) result in wider intervals. Population standard deviation: Larger standard deviations lead to wider intervals. 5. Common Applications Estimating population means, proportions, and differences between groups. Determining the margin of error in surveys and polls. Making inferences about population parameters based on sample data. 6. Important Considerations Confidence intervals do not predict the exact value of the population parameter. The choice of confidence level depends on the desired level of certainty. It's essential to consider the assumptions underlying the confidence interval calculation (e.g., normality, independence). #Confidence #Interval

1

Explore the transformative impact of structured data on Generative AI (GenAI) with our comprehensive study led by Kurt Janssens, AI Product Owner, and Jonas De Keuster, VP of Product Marketing at VaultSpeed. 🚀 🤖 Discover how optimized data organization enhances GenAI performance, reducing errors and improving response accuracy. Our research, featuring ChatGPT-4 experiments and a Retrieval-Augmented Generation (RAG) setup, reveals critical insights for businesses navigating AI-driven decision-making. Gain actionable strategies to leverage structured data effectively and unlock GenAI's full potential. 🔑 🔓 Click to read more: https://lnkd.in/eqAJxEMZ

33

1 commentaar

We're thrilled to unveil a Paradime's latest feature: AI-generated ERDs for dbt™ models! In just a few simple commands, you can now visualize your dbt™ models and their relationships directly within the Paradime IDE. Whether you’re designing complex data architectures or just getting started, this feature help you visualize your data models effortlessly. See tutorial below 👇 #dbt #AnalyticsEngineering #AI #MermaidJS #ERD

16

This one is handy! Used it before to make an index measure. #sql #analytics

1

Our customer report that finally, their data scientists can focus on their job as QuasarDB: ▫️ Reduces development effort: Offers built-in features like FFT and statistical analysis, eliminating the need for custom programming. ▫️ Simplifies model development: Provides a platform for storing both historical and live data, streamlining the model development process. ▫️ Increases data scientist productivity: Frees up valuable time for data scientists to focus on deeper analysis and model improvement. 🌐https://quasar.ai/ #datascience #productivity #modeldevelopment

2

Tired of your boring, London-based Data Job? Here's how to escape it: Don't spam 100s of applications. Focus on higher, fewer-quality ones. When I found my current job, I made 10 applications in total. That's the first step. But how do you actually make these applications good? I've written an email showing you how I did it. It's dropping 17:09 UK time today. You can get the other 2 secrets to finding Data Science jobs below (it's free) https://lnkd.in/eTAmRsDu

5

2 commentaren

Tricks I use to optimise SQL queries: ➡️ SELECT only specific column you need instead of SELECT * ➡️ Avoiding DISTINCT when not necessary. It can be computationally expensive, specially on large datasets ➡️ UNION ALL instead of UNION when you are fine with duplication of rows or you know that there are no duplicates ➡️ Not using ORDER BY until the final SELECT statement Your subqueries and CTEs typically don't need to be ordered. There are many more ways, which I am still not aware of. So what tricks do you use? Comment section is all yours :) ----------------------------------------------------------------- Everyday, I learn and share stuff on: - Data Analytics📊 - Data Science👨🏻‍💻 - ML/MLOps🤖 - Python🐍 If you are interested in above topics, find me here -> Sudhanshu Tiwari

16

🚀 Day 19: Triggers - Automating Database Tasks 🔄✨ Welcome to Day 19 of our SQL series! Today, we’re exploring triggers, a powerful feature in SQL that allows you to automate tasks and enforce business rules in your database. 📚 Why Are Triggers Important? 1. Automate Tasks: Automatically perform tasks such as logging, auditing, and enforcing business rules. 2. Maintain Data Integrity: Ensure data consistency and integrity by validating data changes. 3. Enforce Business Rules: Apply complex business logic consistently across your database. 4. Simplify Maintenance: Reduce the need for manual intervention by automating repetitive tasks. 5. Create Audit Trails: Maintain detailed logs of data changes for security and compliance. Stay tuned for tomorrow’s post: "Data Integrity: Constraints and Keys." If you find this post helpful, please Like and Share it with your network. Your support keeps me motivated to continue sharing valuable content. Thank you! 🙌 #SQL #DataAutomation #DatabaseManagement #Triggers #LearningJourney #LinkedInLearning Shashank Singh 🇮🇳 Aditya Sharma Rishabh Mishra Munna Das Chandeep Chhabra Ankit Bansal Shail Sahu Shakra Shamim Akash Kamerkar Naveen Kumar Adapala Ayan Khan Zain Ul Hassan

45

Dive deep into two of the hot topics at Matillion this week 👇 💡 Migrating from a Matillion Hosted Git Repo to a BYOG repository https://lnkd.in/ewt9jH5N 💡 Medallion Data Architecture for Life Science Data - Databricks Quickstart with Matillion https://lnkd.in/eSvB9f47 Want to how to get more from Matillion and pick up some data engineering tips along the way? www.matillion.com/blog #DataEngineering #AI #Databricks #Git #LifeScience #Medallion #BYOG

9

Phi-3 indeed came with some improvement. I don't trust benchmark data but yeah, that acts as one of my initial filters for selecting a base model. I'm talking about it working on MY specific data. LLaMA-3 also was an improvement over Phi-2 but the improvement was just 2% so given 3x the size, 2% didn't really matter. Now this one, where every model failed to match Phi-2 for me, the new version came with 5.5% improvement on worst examples. I'm using Sequence Classification BTW. Didn't notice much difference in Causal Setting. Now coming to couple of things I found interesting in report, which the Autobots might not haven told you because Transformers can't read between lines, yet. 1. Use can use ## before the question to make it perform better. It's not me but the report says itself in a little tiny corner. 2. They trained this model on Phases. So this thing is intriguing for me as well as I'm kinda sceptic about dataset. They say that "Phase-1 comprises mostly of web sources aimed at teaching the model general knowledge and language understanding. Phase-2 merges even more heavily filtered webdata (a subset used in Phase-1) with some synthetic data that teach the model logical reasoning and various niche skills." They use data filtering, data curation terms a lot but haven't said a single word about the process itself. I'm very very curious about how they could filter the data? Was it manual? Was it automated? If they used LLM generated and Web scraped data, how could they make sure that it is "filtered" to the standards and exact needs. Also, one interesting thing is that Phase-1 data is "aimed at teaching reasoning". Now what that data type could be? How did they make sure that it indeed is aimed at reasoning only. I was more than eager to know so I pinged a few of the researchers mentioned in the paper and they politely refused to answer that. I can totally understand as I too have to watch those mandatory corporate ethics videos every month 🤦‍♂️ If anyone has any ideas on it, please DO let me know. Nothing special in the report except my sceptic annotations: https://lnkd.in/d6rBCUjd #phi #microsoft #phi3 #llm #genai #nlp #ml #ai #research #opensource

87

13 commentaren