Skip to content

clvoloshin/constrained_batch_policy_learning

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

198 Commits
experimental_results
experimental_results
 
 
experimental_results_car
experimental_results_car
 
 
models
models
 
 
seed_2_data
seed_2_data
 
 
tests
tests
 
 
DQN.py
DQN.py
 
 
Pipfile
Pipfile
 
 
README.md
README.md
 
 
car_constraint_values_wo_band.png
car_constraint_values_wo_band.png
 
 
car_main_value.png
car_main_value.png
 
 
car_main_value_wo_band.png
car_main_value_wo_band.png
 
 
car_racing.py
car_racing.py
 
 
config_car.py
config_car.py
 
 
config_lake.py
config_lake.py
 
 
env_dqns.py
env_dqns.py
 
 
env_nn.py
env_nn.py
 
 
exact_policy_evaluation.py
exact_policy_evaluation.py
 
 
exponentiated_gradient.py
exponentiated_gradient.py
 
 
fitted_algo.py
fitted_algo.py
 
 
fitted_off_policy_evaluation.py
fitted_off_policy_evaluation.py
 
 
fittedq.py
fittedq.py
 
 
fixed_policy.py
fixed_policy.py
 
 
fqe_quality_test.py
fqe_quality_test.py
 
 
fqe_quality_test_generalization.py
fqe_quality_test_generalization.py
 
 
fqi_grid_search.py
fqi_grid_search.py
 
 
fqi_seed_2_new.py
fqi_seed_2_new.py
 
 
frozen_lake.py
frozen_lake.py
 
 
inverse_propensity_scoring.py
inverse_propensity_scoring.py
 
 
lake_primal_dual_gap.png
lake_primal_dual_gap.png
 
 
lake_values.png
lake_values.png
 
 
lake_values_wo_band.png
lake_values_wo_band.png
 
 
layer_visualizer.py
layer_visualizer.py
 
 
mdp_approximator.py
mdp_approximator.py
 
 
model.py
model.py
 
 
neural_network.py
neural_network.py
 
 
optimization_problem.py
optimization_problem.py
 
 
pi_old_car_cnn_main.hdf5
pi_old_car_cnn_main.hdf5
 
 
play_car_racing.py
play_car_racing.py
 
 
plot_fqe_quality_test.py
plot_fqe_quality_test.py
 
 
plot_grid_search.py
plot_grid_search.py
 
 
plot_policy_improvement.py
plot_policy_improvement.py
 
 
plot_policy_improvement_v2.py
plot_policy_improvement_v2.py
 
 
plot_results.py
plot_results.py
 
 
print_policy.py
print_policy.py
 
 
replay_buffer.py
replay_buffer.py
 
 
run.py
run.py
 
 
stochastic_policy.py
stochastic_policy.py
 
 
thread_safe.py
thread_safe.py
 
 
value_function.py
value_function.py
 
 

Repository files navigation

constrained_batch_policy_learning

*Note: Use the --headless flag if using a server without a display.

Otherwise, to run the main algorithm:

pip install pipenv
pipenv install
pipenv run python run.py -env car --headless

or, for lake,

pipenv run python run.py -env lake --headless

About

No description, website, or topics provided.

Resources

Readme
Activity

Stars

26 stars

Watchers

3 watching

Forks

11 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages