Adds an _arm_ to the list of arms.
Incorporates the arm into the learning and neighborhood policies with no training data.
Fits the multi-armed bandit to the given decisions, their corresponding rewards and contexts, if any.
Validates arguments and raises exceptions in case there are violations.
@@ -944,7 +1024,7 @@Creates named tuple of the learning policy based on the implementor.
Creates named tuple of the neighborhood policy based on the implementor.
Updates the multi-armed bandit with the given decisions, their corresponding rewards and contexts, if any.
Validates arguments and raises exceptions in case there are violations.
@@ -1006,7 +1086,7 @@Returns the “best” arm (or arms list if multiple contexts are given) based on the expected reward.
The definition of the best depends on the specified learning policy. Contextual learning policies and neighborhood policies require contexts data in training. @@ -1031,7 +1111,7 @@
Returns a dictionary of arms (key) to their expected rewards (value).
Contextual learning policies and neighborhood policies require contexts data for expected rewards.
Removes an _arm_ from the list of arms.
Warm-start untrained (cold) arms of the multi-armed bandit.
Validates arguments and raises exceptions in case there are violations.
The warm-start procedure depends on the learning and neighborhood policy. Note that for certain neighborhood @@ -1109,11 +1189,11 @@
Bases: tuple
Bases: tuple
Clusters Neighborhood Policy.
Clusters is a k-means clustering approach that uses the observations
@@ -1121,7 +1201,7 @@
Supports KMeans and MiniBatchKMeans.
The number of clusters. Integer. Must be at least 2. Default value is 2.
Boolean flag to use MiniBatchKMeans or not. Default value is False.
Bases: tuple
KNearest Neighborhood Policy.
KNearest is a nearest neighbors approach that selects the k-nearest observations to be used with a learning policy.
The number of neighbors to select. Integer value. Must be greater than zero. Default value is 1.
@@ -1189,7 +1269,7 @@The metric used to calculate distance.
Accepts any of the metrics supported by scipy.spatial.distance.cdist.
Default value is Euclidean distance.
Bases: tuple
Locality-Sensitive Hashing Approximate Nearest Neighbors Policy.
LSHNearest is a nearest neighbors approach that uses locality sensitive hashing with a simhash to @@ -1250,7 +1330,7 @@ operation is O(2**n_dimensions).
The number of dimensions to use for the hyperplane. Integer value. Must be greater than zero. Default value is 5.
@@ -1263,7 +1343,7 @@The number of hash tables. Integer value. Must be greater than zero. Default value is 3.
@@ -1276,7 +1356,7 @@The probabilities associated with each arm. Used to select random arm if context has no neighbors. If not given, a uniform random distribution over all arms is assumed. The probabilities should sum up to 1.
@@ -1301,19 +1381,19 @@Alias for field number 2
Bases: tuple
Radius Neighborhood Policy.
Radius is a nearest neighborhood approach that selects the observations within a given radius to be used with a learning policy.
The maximum distance within which to select observations. Integer or Float. Must be greater than zero. Default value is 1.
@@ -1341,7 +1421,7 @@The metric used to calculate distance. Accepts any of the metrics supported by scipy.spatial.distance.cdist. Default value is Euclidean distance.
@@ -1354,7 +1434,7 @@The probabilities associated with each arm. Used to select random arm if context has no neighbors. If not given, a uniform random distribution over all arms is assumed. The probabilities should sum up to 1.
@@ -1379,19 +1459,19 @@Alias for field number 2
Bases: tuple
TreeBandit Neighborhood Policy.
This policy fits a decision tree for each arm using context history. @@ -1416,7 +1496,7 @@ A Practical Method for Solving Contextual Bandit Problems Using Decision Trees, UAI 2017
Parameters of the decision tree. The keys must match the parameters of sklearn.tree.DecisionTreeRegressor. When a parameter is not given, the default parameters from @@ -1443,7 +1523,7 @@
@@ -1453,11 +1533,11 @@This module provides a simulation utility for comparing algorithms and hyper-parameter tuning.
Bases: object
Multi-Armed Bandit Simulator.
This utility runs a simulation using historic data and a collection of multi-armed bandits from the MABWiser @@ -1474,7 +1554,7 @@
A log of the simulation tracks the experiment progress.
A list of tuples of the name of each bandit and the bandit object.
The complete decision history to be used in train and test.
The complete array history to be used in train and test.
The complete context history to be used in train and test.
A scaler object from sklearn.preprocessing.
The size of the test set
Whether to use a chronological division for the train-test split. If false, uses sklearn’s train_test_split.
The size of each batch for online learning.
The function for evaluating the bandits. Values are stored in bandit_to_arm_to_stats_avg. Must have the function signature function(arms_to_stats_train: dictionary, predictions: list, decisions: np.ndarray, rewards: np.ndarray, metric: str).
@@ -1576,7 +1656,7 @@Flag to skip neighborhood statistics.
The logger object.
The list of arms used by the bandits.
Descriptive statistics for the complete data set.
Descriptive statistics for the training data.
Descriptive statistics for the test data.
Descriptive statistics for the predictions made by each bandit based on means from training data.
Descriptive statistics for the predictions made by each bandit based on minimums from training data.
Descriptive statistics for the predictions made by each bandit based on maximums from training data.
The confusion matrices for each bandit.
The prediction for each item in the test set for each bandit.
The arm_to_expectations for each item in the test set for each bandit. For context-free bandits, there is a single dictionary for each batch.
The number of neighbors in each neighborhood for each row in the test set. Calculated when using a Radius neighborhood policy, or a custom class that inherits from it. Not calculated when is_quick is True.
@@ -1722,7 +1802,7 @@The arm_to_stats for each neighborhood for each row in the test set. Calculated when using Radius or KNearest, or a custom class that inherits from one of them. Not calculated when is_quick is True.
@@ -1735,7 +1815,7 @@The indices of the rows in the test set. If input was not zero-indexed, these will reflect their position in the input rather than actual index.
Calculates descriptive statistics for each arm in the provided data set.
Calculates descriptive statistics for the given array of rewards.
Generates a plot of the cumulative sum of the rewards for each bandit. Simulation must be run before calling this method.
Run simulator
Runs a simulation concurrently for all bandits in the bandits list.
Default evaluation function.
Calculates predicted rewards for the test batch based on predicted arms. When the predicted arm is the same as the historic decision, the historic reward is used. @@ -1877,29 +1957,29 @@
This module provides a number of constants and helper functions.
Arm type is defined as integer, float, or string.
alias of Union[int, float, str]
Bases: tuple
Constant values used by the modules.
The distance metrics supported by neighborhood policies.
Num type is defined as integer or float.
alias of Union[int, float]
Returns the first key with the maximum value.
Returns the first key that has the minimum value.
Checks that given expression is false, otherwise raises the given exception.
Checks that given expression is true, otherwise raises the given exception.