Creating a Multiple Dictionaries from a CSV File-CodePudding

I am currently importing a file as so:

df= pd.read_csv(r"Test.csv")

And the output looks like

  Type          Value
0  Food_Place_1   1
1  Food_Place_2   2
2  Car_Type_1     3
3  Car_Type_2     4

I would like to iterate through this df and depending on the Type column allocated to a dictionary like this

food_type_dict = {'Type': ['Food_Place_1', 'Food_Place_2'], 'Value': [1, 2]}

car_type_dict = {'Type': ['Car_Type_1', 'Car_Type_2'], 'Value': [3, 4]}

My plan was to convert the entire dataframe into a dictionary and filter from there. However when I try to convert using this, the output is not what I was expecting. I can't seem to remove the Value header from the dictionary

df_dict = df.set_index(['Type']).T.to_dict('dict')

Output    
{'A1': {'Value': 1},....}

CodePudding user response：

Create category for possible aggregate lists for nested dictionary:

#If category is set by remove digits
cat = df['Type'].str.replace('\d','')

#If category is set by first letter
#cat = df['Type'].str[0]

d = df.rename(columns={'Type':'Component'}).groupby(cat).agg(list).to_dict('index')
print (d)
{'A': {'Component': ['A1', 'A2'], 'Value': [1, 2]}, 
 'B': {'Component': ['B1', 'B2'], 'Value': [3, 4]}}

Then instead a_type_dict use d['A'], b_type_dict use d['B'].

EDIT:

cat = df['Type'].str.split('_').str[0]

d = df.rename(columns={'Type':'Component'}).groupby(cat).agg(list).to_dict('index')
print (d)
{'Car': {'Component': ['Car_Type_1', 'Car_Type_2'], 'Value': [3, 4]}, 
 'Food': {'Component': ['Food_Place_1', 'Food_Place_2'], 'Value': [1, 2]}}