Section: Introduction | Data Visualization on Seaborn | Moodle UM

Section outline

- Select activity Introduction
  
  Seaborn can be used with different types of data, whether Python lists, NumPy arrays, or pandas DataFrames, although pandas DataFrames are generally preferred.
  
  There are different formats of data tables:
  
  Wide-format :
  
  Var1 Val1 Val2 Val3
  
  Var2
  
  Val1 Var3Val11 Var3Val12 Var3Val13
  
  Val2 Var3Val21 Var3Val22 Var3Val23
  
  Val3 Var3Val31 Var3Val32 Var3Val33
  
  Long-format :
  
  Var1 Var2 Var3
  
  Observation1 Var1Val1 Var2Val1 Var3Val1
  
  Observation2 Var1Val2 Var2Val2 Var3Val2
  
  Observation3 Var1Val3 Var2Val3 Var3Val3
  
  The classic format is the long format, which allows you to have data points with many different variables. Here is a description of a table in this format:
  
  It can be useful to check whether any data is missing:
  
  data=sns.load_dataset("penguins") print(data.isnull())#sur le tableau entier print(data.isnull().any())#sur chaque colonne
  
  Then, if you don’t want to include observations with null values for a given variable, for example:
  
  data.dropna(subset=["body_mass_g"])