I have a dictionary like this:
my_dict = {'Southampton': '33.7%', 'Cherbourg': '55.36%', 'Queenstown': '38.96%'}
How can I have a simple plot with 3 bars showing the values of each key in a dictionary?
I've tried:
sns.barplot(x=my_dict.keys(), y = int(my_dict.values()))
But I get :
TypeError: int() argument must be a string, a bytes-like object or a number, not 'dict_values'
There are several issues in your code:
my_dict.values() returns all values as a dict_values object. int(my_dict.values())) means converting the set of all values to a single integer, not converting each of the values to an integer. The former, naturally, makes no sense.float("12.34%"[:-1]).my_dict.keys() and my_dict.values() are not guaranteed to return keys and values in the key-value pairs in the same order, for example, the keys you get may be ['Southampton', 'Cherbourg', 'Queenstown'] and the values you get may be "55.36%", "33.7", "38.96%".With all these issues fixed:
keys = list(my_dict.keys())
# get values in the same order as keys, and parse percentage values
vals = [float(my_dict[k][:-1]) for k in keys]
sns.barplot(x=keys, y=vals)
You get:
my_dict.keys() and my_dict.values() are guaranteed to return the same order. This was also unofficially implemented in C-python 3.6 (not sure about Iron Python or PyPy)
You need to convert the values to numeric, right now they are strings:
import seaborn as sns
my_dict = {'Southampton': '33.7%', 'Cherbourg': '55.36%', 'Queenstown': '38.96%'}
perc = [float(i[:-1]) for i in my_dict.values()]
sns.barplot(x=list(my_dict.keys()),y=perc)
I did the following:
I first removed the % sign from the dict.
my_df = pd.DataFrame(my_dict.items())
ax = sns.barplot(x=0, y=1, data=my_df)
ax.set(xlabel = 'Cities', ylabel='%', title='Title')
