I try to discretize some numbers, by looking if they are in a given range, and then assign a number based on the range, however the result which I get is not exactly correct.
mapp is the a dictionary of which defines ranges, and the values which correspond to the given range.
lst is the list of numbers that I want to match against those ranges, and assign identifiers to them
mapp = {(0,100): 1, (100,400): 2, (400,800): 3}
lst = [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]
def discretize(mapping_dict, list_of_values):
print "\n"
location = []
for x in sorted(list_of_values):
for (lower_bound,upper_bound),value in mapping_dict.items():
if round(x) in range(lower_bound,upper_bound):
print round(x), "yes", value
distance = mapping_dict[(lower_bound,upper_bound)]
location.append((distance))
else:
print round(x), "no"
distance = len(mapping_dict.items())+10
location.append((distance))
return location
The result which I expect is: [1, 1, 2, 3, 3, 13] , however that's not what I get.
This is the actual result which I get, which is incorrect:
4.0 yes 1
4.0 no #wrong!
5.0 yes 1
5.0 no #wrong!
300.0 yes 2
300.0 no #wrong!
501.0 yes 3
501.0 no #wrong!
600.0 yes 3
600.0 no #wrong!
901.0 no #CORRECT
[1, 13, 1, 13, 2, 13, 3, 13, 3, 13, 13]
I get no at 4.0 which is not correct, etc, etc.
Where is the problem?
Thanks
mapp = {(0,100): 1, (100,400): 2, (400,800): 3}
lst = [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]
result = []
for l in lst:
for m in mapp:
if m[0] < l < m[1]:
result.append(mapp[m])
print result
Output:
[1, 1, 2, 3, 3]
EDIT:
result = []
for l in lst:
flag=True
for m in mapp:
if m[0] < l < m[1]:
result.append(mapp[m])
flag = False
break
if flag:
result.append(-1)
print result
Output:
[1, 1, 2, 3, 3, -1]
I think I have faced a similar problem once, because I found a small RangeDict class:
class RangeDict (dict):
def __init__ (self, *args):
super ().__init__ ()
def __setitem__ (self, k, v):
if not isinstance (k, slice): raise ValueError ('Indices must be slices.')
super ().__setitem__ ( (k.start, k.stop), v)
def __getitem__ (self, k):
for (start, stop), v in self.items ():
if start <= k < stop: return v
raise IndexError ('{} out of bounds.'.format (k) )
I hope this class wraps your desired funcionality. Obviously lookup is O(N) and not O(1).
Sample usage:
r = RangeDict ()
r [0:100] = 1
r [100:400] = 2
r [400:800] = 3
for x in [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]:
print (r [x] )
#Last value raises IndexError
Putting an else after your for loop you were an the right track! When you put an else after a loop, that else block is executed each time the loop exits normally, i.e. without using e.g. break. Thus, (assuming that your groups are non-overlapping) you just need to add a break statement to the end of your if block, i.e. after location.append((distance)). Then it works as expected.
Also, instead of checking whether the number is in the range (which creates and searches a list each time!) you should just use <= and <. Also, you already have the value, so why not use it?
for (lower_bound, upper_bound), value in mapping_dict.items():
if lower_bound <= x < upper_bound:
location.append(value)
break
else:
location.append(len(mapping_dict) + 10)
I try to discretize some numbers, by looking if they are in a given range, and then assign a number based on the range, however the result which I get is not exactly correct.
mapp is the a dictionary of which defines ranges, and the values which correspond to the given range.
lst is the list of numbers that I want to match against those ranges, and assign identifiers to them
mapp = {(0,100): 1, (100,400): 2, (400,800): 3}
lst = [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]
def discretize(mapping_dict, list_of_values):
print "\n"
location = []
for x in sorted(list_of_values):
for (lower_bound,upper_bound),value in mapping_dict.items():
if round(x) in range(lower_bound,upper_bound):
print round(x), "yes", value
distance = mapping_dict[(lower_bound,upper_bound)]
location.append((distance))
else:
print round(x), "no"
distance = len(mapping_dict.items())+10
location.append((distance))
return location
The result which I expect is: [1, 1, 2, 3, 3, 13] , however that's not what I get.
This is the actual result which I get, which is incorrect:
4.0 yes 1
4.0 no #wrong!
5.0 yes 1
5.0 no #wrong!
300.0 yes 2
300.0 no #wrong!
501.0 yes 3
501.0 no #wrong!
600.0 yes 3
600.0 no #wrong!
901.0 no #CORRECT
[1, 13, 1, 13, 2, 13, 3, 13, 3, 13, 13]
I get no at 4.0 which is not correct, etc, etc.
Where is the problem?
Thanks
mapp = {(0,100): 1, (100,400): 2, (400,800): 3}
lst = [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]
result = []
for l in lst:
for m in mapp:
if m[0] < l < m[1]:
result.append(mapp[m])
print result
Output:
[1, 1, 2, 3, 3]
EDIT:
result = []
for l in lst:
flag=True
for m in mapp:
if m[0] < l < m[1]:
result.append(mapp[m])
flag = False
break
if flag:
result.append(-1)
print result
Output:
[1, 1, 2, 3, 3, -1]
I think I have faced a similar problem once, because I found a small RangeDict class:
class RangeDict (dict):
def __init__ (self, *args):
super ().__init__ ()
def __setitem__ (self, k, v):
if not isinstance (k, slice): raise ValueError ('Indices must be slices.')
super ().__setitem__ ( (k.start, k.stop), v)
def __getitem__ (self, k):
for (start, stop), v in self.items ():
if start <= k < stop: return v
raise IndexError ('{} out of bounds.'.format (k) )
I hope this class wraps your desired funcionality. Obviously lookup is O(N) and not O(1).
Sample usage:
r = RangeDict ()
r [0:100] = 1
r [100:400] = 2
r [400:800] = 3
for x in [3.5, 5.4, 300.12, 500.78, 600.45, 900.546]:
print (r [x] )
#Last value raises IndexError
Putting an else after your for loop you were an the right track! When you put an else after a loop, that else block is executed each time the loop exits normally, i.e. without using e.g. break. Thus, (assuming that your groups are non-overlapping) you just need to add a break statement to the end of your if block, i.e. after location.append((distance)). Then it works as expected.
Also, instead of checking whether the number is in the range (which creates and searches a list each time!) you should just use <= and <. Also, you already have the value, so why not use it?
for (lower_bound, upper_bound), value in mapping_dict.items():
if lower_bound <= x < upper_bound:
location.append(value)
break
else:
location.append(len(mapping_dict) + 10)
0 commentaires:
Enregistrer un commentaire