lcc.stars_processing.utilities package¶
Submodules¶
lcc.stars_processing.utilities.base_decider module¶
-
class
lcc.stars_processing.utilities.base_decider.
BaseDecider
[source]¶ Bases:
object
A decider class works with “coordinates” (specification) of objects. It can learn identify inspected group of objects according to “coordinates” of searched objects and other objects.
All decider classes have to inherit this abstract class. That means that they need to implement several methods: “learn” and “evaluate”. Also all of them have to have “treshold” attribute. To be explained read comments below.
Attributes
treshold (float) Probability (1.0 means 100 %) level. All objects with probability of membership to the group higher then the treshold are considered as members. treshold = 0.8 Methods
-
evaluate
(star_coords)[source]¶ Parameters: star_coords : list
Coordinates of inspected star got from sub-filters
Returns: list of lists
Probability that inspected star belongs to the searched group of objects
-
evaluateList
(stars_coords)[source]¶ Parameters: stars_coords : list
Coordinates of inspected stars (e.g. obtained from sub-filters)
Returns: list
Probabilities that inspected stars belongs to the searched group of objects
-
filter
(stars_coords, treshold=None)[source]¶ Parameters: stars_coords : list
Coordinates of inspected stars
treshold : float
Treshold value for filtering (number from 0 to 1)
Returns: List of True/False whether coordinates belong to the searched group of objects
-
getBestCoord
(stars_coords)[source]¶ Parameters: stars_coords : list
Coordinates of inspected stars got from sub-filters
Returns: list
Coordinates with highest probability of membership to the searched group (one list of coordinates)
-
getStatistic
(right_coords, wrong_coords, treshold=None)[source]¶ Parameters: right_coords : list
Parameter-space coordinates of searched objects
wrong_coords : list
Parameter-space coordinates of other objects
treshold : float
Treshold value for filtering (number from 0 to 1)
Returns: statistic information : dict
- precision (float)
True positive / (true positive + false positive)
- true_positive_rate (float)
Proportion of positives that are correctly identified as such
- true_negative_rate :(float)
Proportion of negatives that are correctly identified as such
- false_positive_rate (float)
Proportion of positives that are incorrectly identified as negatives
- false_negative_rate (float)
Proportion of negatives that are incorrectly identified as positives
-
learn
(right_coords, wrong_coords)[source]¶ After executing this method the decider object is capable to recognize objects according their “coordinates” via “filter” method.
Parameters: right_coords : list
“Coordinates” of searched objects
wrong_coords : list
“Coordinates” of other objects
Returns: NoneType
None
-
lcc.stars_processing.utilities.base_descriptor module¶
-
class
lcc.stars_processing.utilities.base_descriptor.
BaseDescriptor
[source]¶ Bases:
object
Methods
-
LABEL
= ''¶
-
-
class
lcc.stars_processing.utilities.base_descriptor.
Learnable
[source]¶ Bases:
object
Common class for all filters which are able to call “learn” by yourself. All these classes need to be able obtain their space coordinates via getSpaceCoords. Then the learning is the same (see learn method below).
Optionally there can be labels on plots if a class has label attribute, which is list of string contains label for data.
Also after learning the ‘learned’ attribute is set to ‘True’ if exists.
Moreover plot is saved if class has plot_save_path attribute is not None or ‘’
Methods
-
getSpaceCoords
(stars)[source]¶ Parameters: stars : list of Star objects
Returns: list of lists
List of list of numbers (coordinates)
-
learn
(searched_stars, contamination_stars, learn_num='')[source]¶ Teach filter to recognize searched stars
Parameters: searched_stars : list of Star objects
Searched stars to learn
contamination_stars : list of Star objects
Contamination stars to learn
learn_num : str, int
Optional identifier for the learning
Returns: None
-
lcc.stars_processing.utilities.compare module¶
-
class
lcc.stars_processing.utilities.compare.
ComparativeBase
[source]¶ Bases:
object
This class is responsible for comparing light curves of inspected stars with the template stars
Attributes
compar_stars (list, iterable) List of Star objects which represent searched group of star objects Methods
lcc.stars_processing.utilities.sax module¶
-
exception
lcc.stars_processing.utilities.sax.
DictionarySizeIsNotSupported
[source]¶ Bases:
exceptions.ValueError
-
exception
lcc.stars_processing.utilities.sax.
OverlapSpecifiedIsNotSmallerThanWindowSize
[source]¶ Bases:
exceptions.ValueError
-
class
lcc.stars_processing.utilities.sax.
SAX
(word_size=8, alphabet_size=10, scaling_factor=1)[source]¶ Bases:
object
This class manages symbolic representation of data series via Symbolic Aggregate approXimation method. It translates series of data to a words, which can then be compared with other such words in symbolic distance space.
Attributes
word_size (int) Number of letters in transformed word alphabet_size (int) Size of alphabet counted from A (3 means A, B, C) scaling_factor (int, float) Scaling factor can be used to scale result dissimilarity of two words created from light curves of different lengths beta (list) Breakpoints for given alphabets size Methods
-
A_OFFSET
= 97¶
-
MAX_ALPH_SIZE
= 20¶
-
MIN_ALPH_SIZE
= 3¶
-
alphabetize
(paaX)[source]¶ Converts the Piecewise Aggregate Approximation of x to a series of letters.
- paaX : list, iterable
- Data series (list of numbers)
Returns: str
SAX word
-
build_letter_compare_dict
()[source]¶ Builds up the lookup table to determine numeric distance between two letters given an alphabet size.
Returns: None
-
compare_letters
(la, lb)[source]¶ Compare two letters based on letter distance return distance between
- la : str
- First letter
- lb : str
- Second letter
Returns: float
Distance between two letters
-
lcc.stars_processing.utilities.superv_base_decider module¶
-
class
lcc.stars_processing.utilities.superv_base_decider.
SupervisedBase
(clf, treshold=0.5)[source]¶ Bases:
lcc.stars_processing.utilities.base_decider.BaseDecider
Base class for sklearn library supervised classes transformed to the package content. It is not intended to use this directly, but thru certain method subclasses.
Attributes
treshold (float) Border probability value (objects with probability higher then this value is considered as searched object) learner (sklearn object) Learner object for desired method of supervised learning Methods
lcc.stars_processing.utilities.symbolic_representation module¶
lcc.stars_processing.utilities.unsupervised_base module¶
Module contents¶
There are managers responsible for stars to have certain parameters for filtering