Related Posts
Conversation Starter
Community Builder
More Posts
Especially for companies doing heavy hiring!
Community Builder
Rising Star
Is demand forecasting still a problem to be solved in today's FMCG industry? Or are all the possible solutions already there? We can always innovate and improve our forecasting models but just wanted to understand the market. HSBC India Tata Consultancy Accenture Infosys WNS Global Services EY Deloitte PwC KPMG
Additional Posts in Data & Analytics Consultants
Mentor
Anyone good in MySQL?? I’m struggling
New to Fishbowl?
Download the Fishbowl app to
unlock all discussions on Fishbowl.
unlock all discussions on Fishbowl.
why are you worried about not developing SQL skills..? Instead the worry should be am I good at coding in Python? Because that’s where the focus is now.
I’m learning Python now, too. I just have a lot more experience in SQL and want to push the envelope there. I think SQL and Python complement each other nicely.
I like doing some cleaning/manipulation in SQL and then throwing it into a temp table that I can read into Python for further analysis, vs. doing the cleansing in Python.
Gotcha.. would recommend doing the cleansing work in Python for three reasons:
1. We are headed in a direction where queries would be written and executed in Python
2. It’s efficient and would improve your coding skill set
3. It is fun to cleanse data in Python
As to SQL.. if you are good at it, don’t worry about how much more you can learn. If you really want to go advanced do PL/SQL programming. There are components to handle large volumes of data in PL/SQL
@OP the next step from writing SQL is improving the database so your SQL runs better. Then it's writing python that can be dropped in any environment and write the SQL for you. Then proper release process, user management, and so on. If you're just querying it's like saying you've peaked at driving the car without knowing how to get more performance out of the engine.
Yeah SM1 has it exactly - automate the data prep step. Have something check keys for nulls/blanks. Build scripts that can validate unseen tables in minutes. Makes life much easier and can save you pain down the line if you missed issues.
I work in Python. I can write a SQL query when I need to but it's not something I feel the need to get better at right now. I can get the data I need to get - the next steps are doing it more efficiently, but that's not going to help me right now.
For me it's like regular expressions - I need to be able to use it from time to time, but there are so many online resources and memorizing a lot of specifics doesn't make sense. It's not interesting conceptually. It's not going to impress anyone for my next job because they assume (correctly) that I can write a SQL query. And I don't want to be in a job where my skill at writing really efficient SQL queries matters.
You can probably do 99% of what you need to with SQL and use python as a procedural wrapper. You get results super fast and everybody can claim “python”