Guidance on improving or learning properly Data Science /Machine Learning
Hi maybe a weird one to ask I graduated in 2017 in MSc Data Science. learned SQL ,R Applied Statistic(Basic ML), Big data Hadoop.
Since then worked as data analyst working with SAP and Dashboards, for 2 years. Then moved to a start up which was good worked with python SQL, did various things building automation pipelines , automation, data auditing, few ML projects, looked into LLM for data cleaning. data migration to AWS and data analytics. did a mix of things.
Then moved to a data science role for recommendation system learned how that works but left after few months due pay being to low. Moved to a big cooperation which is a lot more slow paced. The work is more with a cloud provider and dataform moving data pipelines and data adhoc tasks at the moment and looking at work it will take some time where I b working with ML.
But from my experience I have not done much ML projects in terms of learning
to actually understand what and how it work and what to actually what is a good way to learn. If you don't use something you wont get much experience
How do you know which model to use and which one is the right one?
How do move beyond modeling and build a full end to end ml?
What i struggle with is ok which is the right model how do you evaluate it properly
and what do you after it.
Also how many models should I learn and actually understand?