One of my biggest mistakes when I began learning data science - I believed everything I heard.

Don’t do this. 🧵

#datascience #rstats #python

Here are 2 of the biggest fallacies that cost me time.

I hear that data scientists must be experts in:

👉Math: advanced calculus, matrix math, statistical analysis, the list goes on.

I can count on one hand how many times I’ve used advanced calculus.

Here’s what really helped.

When I started I had a good grasp of basic stats.

Counts, summaries, average, standard deviation, & correlation.

Along the way, I picked up knowledge of:

1. How regression works
2. Histograms, density, & distributions
3. Within-group analysis.

👉Deep Learning: TensorFlow, Keras, PyTorch.

For 99% of problems I don’t use DL.

I really only have had success using deep learning in 1 important area: Time series.

I use ML for everything else.

Here’s what really helped:

Over time, I got good at Machine Learning.

I learned how to use all of the major algorithms:

1. Elastic Net
2. Random Forest
3. XGBOOST
4. KNN
5. SVM

NOW for the TRUTH.

The truth is that all you need to get started is basic understanding of math and motivation.

Everything else can be learned.

Now I will say this:

It took me embarrassingly long (5 years) to become confident in data science.

But I have good news!

I have consolidated 5-years of learning into a free 40-minute webinar that covers:

- 10 secrets I wished I knew going into data science.

- 3 most common mistakes (I suffered from all of these)

- 1 roadmap to success (based on years of experience)

This presentation is potent.

Here's a link: learn.business-science.io/free-rtrack-masterclass-signup

Just watch and take notes.