I am using a package where I need to build a class object for each of my experiments (many) and each time I need to feed too many variables into the class.
the class source code looks like:
.
.
.
def __init__(self, **kwargs):
keys_to_set = [
'key1',
'key2',
'key3',
.
.
'keyN']
.
.
.
Now, I want to create .csv files for each experiment, and feed the class through an automated manner e.g. with a dictionary, however dictionary as arguments does not work for me and get errors!
I'm out of ideas, I don't know if there are any "dictionary-like" structures that may help me here. any helps/tips would be appreciated!
to make it a bit clear maybe:
if I want to make an object like test_obj like this:
test_obj = Class(key1=val1,
key2=val2,
.
.
keyN=valN)
is there any way that I can pass the argument as a dictionary-type thing like:
test_dict = {key1: val1, ...., keyN: valN}
test_obj = Class(test_dict)
(apparently dictionary can not do the job as I have tested)
I think you are looking to do one of these two things:
This creates an instance based completely on passing a dictionary.
class Foo():
def __init__(self, data):
for key, value in data.items():
self.__setattr__(key, value)
foo = Foo({"foo": 1, "bar": 2})
print(foo.foo, foo.bar)
Alternatively, this does the same in this particular case.
class Foo():
def __init__(self, data):
self.__dict__ = data
foo = Foo({"foo": 1, "bar": 2})
print(foo.foo, foo.bar)
Perhaps though, your constructor is less dynamic and you want to pass it key value pairs based on a dictionary. In that case you might:
class Bar():
def __init__(self, foo, bar):
self.foo = foo
self.bar = bar
bar = Bar(**{"foo": 1, "bar": 2})
print(bar.foo, bar.bar)
All examples should result in:
1 2