TigerControl reference documentation¶
For an introduction to TigerControl, start at the TigerControl GitHub page.
tigercontrol package¶
Subpackages¶
tigercontrol.utils package¶
dataset_registry¶
unemployment([verbose]) |
Description: Checks if unemployment data exists, downloads if not. |
uci_indoor([verbose]) |
Description: Checks if uci_indoor data exists, downloads if not. |
sp500([verbose]) |
Description: Checks if S&P500 data exists, downloads if not. |
crypto() |
Description: Checks if cryptocurrency data exists, downloads if not. |
enso(input_signals, include_month, …) |
Description: Transforms the ctrl_indices dataset into a format suitable for online learning. |
random¶
set_key([key]) |
Descripton: Fix global random key to ensure reproducibility of results. |
generate_key() |
Descripton: Generate random key. |
get_global_key() |
Descripton: Get current global random key. |
tigercontrol.problems package¶
custom¶
tigercontrol.problems.CustomProblem() |
Description: class for implementing algorithms with enforced modularity |
tigercontrol.problems.register_custom_problem(…) |
Description: global custom problem method |
control¶
tigercontrol.problems.ControlProblem() |
Description: class for online control tests |
tigercontrol.problems.control.LDS_Control() |
Description: Simulates a linear dynamical system. |
tigercontrol.problems.control.LSTM_Control() |
Description: Produces outputs from a randomly initialized recurrent neural network. |
tigercontrol.problems.control.RNN_Control() |
Description: Produces outputs from a randomly initialized recurrent neural network. |
tigercontrol.problems.control.CartPole() |
Description: |
tigercontrol.problems.control.DoublePendulum() |
Acrobot is a 2-link pendulum with only the second joint actuated. |
tigercontrol.problems.control.Pendulum([g]) |
time_series¶
tigercontrol.problems.TimeSeriesProblem() |
Description: class for online control tests |
tigercontrol.problems.time_series.SP500() |
Description: Outputs the daily opening price of the S&P 500 stock market index from January 3, 1986 to June 29, 2018. |
tigercontrol.problems.time_series.UCI_Indoor() |
Description: Outputs various weather metrics from a UCI dataset from 13/3/2012 to 11/4/2012 |
tigercontrol.problems.time_series.ENSO() |
Description: Collection of monthly values of control indices useful for predicting La Nina/El Nino. |
tigercontrol.problems.time_series.Crypto() |
Description: Outputs the daily price of bitcoin from 2013-04-28 to 2018-02-10 |
tigercontrol.problems.time_series.Random() |
Description: A random sequence of scalar values taken from an i.i.d. |
tigercontrol.problems.time_series.ARMA() |
Description: Simulates an autoregressive moving-average time-series. |
tigercontrol.problems.time_series.Unemployment() |
Description: Monthly unemployment rate since 1948. |
tigercontrol.problems.time_series.LDS_TimeSeries() |
Description: Simulates a linear dynamical system. |
tigercontrol.problems.time_series.LSTM_TimeSeries() |
Description: Produces outputs from a randomly initialized recurrent neural network. |
tigercontrol.problems.time_series.RNN_TimeSeries() |
Description: Produces outputs from a randomly initialized recurrent neural network. |
pybullet¶
tigercontrol.problems.pybullet.PyBulletProblem |
|
tigercontrol.problems.pybullet.Simulator |
|
tigercontrol.problems.pybullet.Ant |
|
tigercontrol.problems.pybullet.CartPole |
|
tigercontrol.problems.pybullet.CartPoleDouble |
|
tigercontrol.problems.pybullet.CartPoleSwingup |
|
tigercontrol.problems.pybullet.HalfCheetah |
|
tigercontrol.problems.pybullet.Humanoid |
|
tigercontrol.problems.pybullet.Kuka |
|
tigercontrol.problems.pybullet.KukaDiverse |
|
tigercontrol.problems.pybullet.Minitaur |
|
tigercontrol.problems.pybullet.Obstacles |
tigercontrol.models package¶
control¶
tigercontrol.models.control.ControlModel() |
Description: class for implementing algorithms with enforced modularity |
tigercontrol.models.control.KalmanFilter() |
Description: Kalman Filter adjusts measurements of a signal based on prior states and knowledge of intrinsic equations of the system. |
tigercontrol.models.control.ODEShootingMethod() |
Description: Implements the shooting method to solve second order boundary value problems with conditions y(0) = a and y(L) = b. |
tigercontrol.models.control.LQR() |
Description: Computes optimal set of actions using the Linear Quadratic Regulator algorithm. |
tigercontrol.models.control.MPPI() |
Description: Implements Model Predictive Path Integral Control to compute optimal control sequence. |
tigercontrol.models.control.CartPoleNN() |
Description: Simple multi-layer perceptron policy, no internal state |
tigercontrol.models.control.ILQR() |
Description: Computes optimal set of actions using the Linear Quadratic Regulator algorithm. |
time_series¶
tigercontrol.models.time_series.TimeSeriesModel() |
Description: class for implementing algorithms with enforced modularity |
tigercontrol.models.time_series.AutoRegressor() |
Description: Implements the equivalent of an AR(p) model - predicts a linear combination of the previous p observed values in a time-series |
tigercontrol.models.time_series.LastValue() |
Description: Predicts the last value in the time series, i.e. |
tigercontrol.models.time_series.PredictZero() |
Description: Predicts the next value in the time series to be 0, i.e. |
tigercontrol.models.time_series.RNN() |
Description: Produces outputs from a randomly initialized recurrent neural network. |
tigercontrol.models.time_series.LSTM() |
Description: Produces outputs from a randomly initialized LSTM neural network. |
tigercontrol.models.time_series.LeastSquares() |
Description: Implements online least squares. |
optimizers¶
tigercontrol.models.optimizers.Optimizer([…]) |
Description: Core class for model optimizers |
tigercontrol.models.optimizers.Adagrad([…]) |
Description: Ordinary Gradient Descent optimizer. |
tigercontrol.models.optimizers.Adam([pred, …]) |
Description: Ordinary Gradient Descent optimizer. |
tigercontrol.models.optimizers.ONS([pred, …]) |
Online newton step algorithm. |
tigercontrol.models.optimizers.SGD([pred, …]) |
Description: Stochastic Gradient Descent optimizer. |
tigercontrol.models.optimizers.OGD([pred, …]) |
Description: Ordinary Gradient Descent optimizer. |
tigercontrol.models.optimizers.mse(y_pred, …) |
Description: mean-square-error loss :param y_pred: value predicted by model :param y_true: ground truth value :param eps: some scalar |
tigercontrol.models.optimizers.cross_entropy(…) |
Description: cross entropy loss, y_pred is equivalent to logits and y_true to labels :param y_pred: value predicted by model :param y_true: ground truth value :param eps: some scalar |
boosting¶
tigercontrol.models.boosting.SimpleBoost() |
Description: Implements the equivalent of an AR(p) model - predicts a linear combination of the previous p observed values in a time-series |
tigercontrol.experiments package¶
core¶
create_full_problem_to_models(problems_ids, …) |
Description: Associate all given problems to all given models. |
run_experiment(problem, model[, metric, …]) |
Description: Initializes the experiment instance. |
metrics¶
mse(y_pred, y_true) |
Description: mean-square-error loss |
cross_entropy(y_pred, y_true[, eps]) |
Description: cross entropy loss, y_pred is equivalent to logits and y_true to labels |
experiment¶
Experiment() |
Description: Experiment class |
new_experiment¶
NewExperiment() |
Description: class for implementing algorithms with enforced modularity |
precomputed¶
recompute([verbose, load_bar]) |
Description: Recomputes all the results. |
load_prob_model_to_result([problem_ids, …]) |
Description: Initializes the experiment instance. |
License¶
Some license