The progress that machine learning has made in past decade strikes everyone as genuine and astounding. Tons of libraries, architectures and mathematical equations have been developed to support machine learning. Such a growth is attracting every developer. No matter even if he has been doing quantum computing, he might want to learn machine learning. It might generate some insightful prediction for him. But the most commonest question I face when I meet new developers is to do with getting started.

After losing in a playoff to make it out of the local qualifying for the 2018 US Open at Shinnecock, I’m stuck at my apartment watching everyone struggle, wondering how much I’d be struggling if I was there myself.

R Tip: be wary of “...“.

The biggest competition in the world’s most popular sport returns this week, and from bars to office pools, predicting the winners and losers will be the topic of countless conversations. And just like in every tournament, these conversations will be dominated by the underdogs that do well and the giants that crash out early.

Welcome to Kaggle Data Notes!

Data are becoming the new raw material of business The Economist

Python generators are a powerful, but misunderstood tool. They’re often treated as too difficult a concept for beginning programmers to learn — creating the illusion that beginners should hold off on learning generators until they are ready. I think this assessment is unfair, and that you can use generators sooner than you think.

The R package wrapr 1.5.0 is now available on CRAN.

In the previous post, we covered the following topics:

In the last model, the embedding matrix was initialized randomly. What if we could use pre-trained word embeddings to intialize it instead?