Comparing Movie Platforms — TMDb vs IMDb
D3 library of JavaScript is known for its flexibility and power for creating interactive and/or animated visualizations and also its steep learning curve.
This library is developed by Mike Bostock who also provided enormous number of high-quality visualizations and their codes. Those examples can be found on the official website here. I found those example visualizations as great resources even for brainstorming to decide what type of visualization could be a good fit for your purpose though, the main power is coming from the codes provided which can be adapted pretty easily after understanding some basics of D3.
Is it really as hard as it looks?
It is…
Validation Methods
Assessing the severity and spotting the causes of dissimilar training and testing set using Airbnb data from Melbourne and Sydney.
One main assumption ML models rely on is that training and validation sets are derived from the same population and identically distributed. This assumption is never totally correct though if training and test sets have completely different distributions, anything your model learn from training data would not be applicable to the test data.
For example, let’s imagine a real estate data of Melbourne and Sydney with columns as follows; “Suburb Name”, “Number of bedrooms” and “Price”. If your training set only have houses in Melbourne, Is it a good idea to keep “Suburb Name” column in the features list?
Given…
Anil Gurbuz
Data Scientist
