Cleaning missing values and invalid data
By default, the pandas read_csv() function will read a variable as if it’s non-numeric (string) if it contains at least one string (text). So, what’s the difference between nan instances in the Petal_width column and the Sepal_width column? Python will convert empty cells into nan values but will keep the numeric nature of the variable, as is the case for the Petal_length variable.
In biostatistics, experimenters might use different words to mark a missing value, such as Nan or NA (short for not applicable), or even whole words such as missing or not applicable. Remember that Nan and NA are still strings, so if there’s an empty cell, Python will read it as a string and coerce the whole variable into a string variable. This wasn’t the case for Petal_width since Python read empty cells as nan and didn’t coerce the variable into a string, instead keeping it numeric. In this case, Python read nan as the valid...
