ResultsDict

class ResultsDict

Bases: UserDict

Class for keeping track of results

__init__(filename, replace=False, lock=True)

initialise empty ResultsDict :param replace: set the replace attribute (default: False)

Methods

__init__	initialise empty ResultsDict :param replace: set the replace attribute (default: False)
clear
contains_dim	equivalent to dim in self['dims']
copy
delete_dim
fromkeys
get
items
keys
load	restore results from file
max	return maximum auc values
min
pop	If key is not found, d is returned if given, otherwise KeyError is raised.
popitem	as a 2-tuple; but raise KeyError if D is empty.
reduce_to_dims	remove all data for dimensions not in dims :param dims: list of dimensions to keep
runs	return the number of runs
save	dump contents to json file
setdefault
update	If E present and has a .keys() method, does: for k in E: D[k] = E[k] If E present and lacks .keys() method, does: for (k, v) in E: D[k] = v In either case, this is followed by: for k, v in F.items(): D[k] = v
update_dim	update data for given dimension
values

Attributes

replace

if True, updates replace existing data, if False, updates append data

load()

restore results from file

Parameters:

• filename – input json file

• replace – set the replace attribute

Returns:

populated ResultsDict

save()

dump contents to json file

Parameters:

filename – output file path

update_dim(dim, replace=False, **kwargs)

update data for given dimension

Parameters:

• dim – dimension to update

• auc – new auc value

• loss – new loss value

• args – new args data (optional)

if self.contains_dim(dim) == True, behaviour depends on the value of self.replace

delete_dim(dim)

max(field, dim=None)

return maximum auc values

Parameters:

• field – field to take maximum over

• dim – if dim=None, return list of values for all dimension, else only return maximum value for dim.

min(field, dim=None)

get(k[, d]) → D[k] if k in D, else d.  d defaults to None.

contains_dim(dim)

equivalent to dim in self['dims']

reduce_to_dims(dims)

remove all data for dimensions not in dims :param dims: list of dimensions to keep

runs(dim=None)

return the number of runs

Parameters:

dim – if dim is None, return list of number of runs for all dimension, else return number of runs for dimension dim.