In general, pandas tries to load and store data efficiently. It automatically assigns data types (which we can inspect by calling the dtypes method of a pandas DataFrame). However, there are some tricks that can lead to much better memory allocation, which definitely make working with larger tables (in hundreds of MBs, or even GBs) easier.
- We start by inspecting the data types in our DataFrame:
df.dtypes
We only show a snippet of all the columns, for brevity:
In the preceding image, we see a few distinct data types: floats (floating-point numbers, such as 3.42), integers, and objects. The last ones are the pandas representation ...
