Skip to content

Instantly share code, notes, and snippets.

@cvigoe

cvigoe/RL_plot.py

Created June 25, 2020 21:59
Show Gist options
  • Save cvigoe/27f60fd176439403795866d938f86878 to your computer and use it in GitHub Desktop.
Save cvigoe/27f60fd176439403795866d938f86878 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
RL_plot.py
NUM_ALGO_TRIALS = _
NUM_TRAINING_TIMESTEPS = _
NUM_TESTING_TIMESTEPS = _
set_seed()
for algo_iteration in range(NUM_ALGO_TRIALS):
for training_timesteps in range(NUM_TRAINING_TIMESTEPS): #for epsiodic tasks, keep resetting the env
deploy_policy_with_exploration()
update_policy_with_new_data()
for testing_timesteps in range(NUM_TESTING_TIMESTEPS): #for epsiodic tasks, keep resetting the env
deploy_policy_without_exploration()
compute_mean_return_from_test_data()
add_mean_return_to_log(algo_iteration,training_episode)
plot_return_mean(log) #for each point on x-axis, compute mean using NUM_ALGO_TRIALS "samples" (av. of averages)
plot_return_SD(log) #for each point on x-axis, compute SD using NUM_ALGO_TRIALS "samples" (SD of averages)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment