Linting

experiments/mpsc/mpsc_experiment.py

@@ -163,7 +163,7 @@ def determine_feasible_starting_points(num_points=100):
     # Define arguments.
     fac = ConfigFactory()
     config = fac.merge()
-    config.sf_config.cost_function='one_step_cost'
+    config.sf_config.cost_function = 'one_step_cost'
 
     task = 'stab' if config.task_config.task == Task.STABILIZATION else 'track'
     if config.task == Environment.QUADROTOR:

experiments/mpsc/plotting_results.py

@@ -14,8 +14,6 @@
 
 plot = False
 save_figs = True
-ordered_algos = ['lqr', 'ppo', 'sac']
-# ordered_algos = ['lqr', 'pid', 'ppo', 'sac']
 
 U_EQs = {
     'cartpole': 0,
@@ -27,51 +25,6 @@
 met.verbose = False
 
 
-def load_one_experiment(system, task, algo, mpsc_cost_horizon):
-    '''Loads the results of every MPSC cost function for a specific experiment.
-
-    Args:
-        system (str): The system to be controlled.
-        task (str): The task to be completed (either 'stab' or 'track').
-        algo (str): The controller being used.
-        mpsc_cost_horizon (str): The cost horizon used by the smooth MPSC cost functions.
-
-    Returns:
-        all_results (dict): A dictionary containing all the results.
-    '''
-
-    all_results = {}
-
-    for cost in ordered_costs:
-        with open(f'./results_mpsc/{system}/{task}/m{mpsc_cost_horizon}/results_{system}_{task}_{algo}_{cost}_cost_m{mpsc_cost_horizon}.pkl', 'rb') as f:
-            all_results[cost] = pickle.load(f)
-
-    return all_results
-
-
-def load_all_algos(system, task, mpsc_cost_horizon):
-    '''Loads the results of every MPSC cost function for a specific experiment with every algo.
-
-    Args:
-        system (str): The system to be controlled.
-        task (str): The task to be completed (either 'stab' or 'track').
-        mpsc_cost_horizon (str): The cost horizon used by the smooth MPSC cost functions.
-
-    Returns:
-        all_results (dict): A dictionary containing all the results.
-    '''
-
-    all_results = {}
-
-    for algo in ['lqr', 'pid', 'ppo', 'sac']:
-        if system == 'cartpole' and algo == 'pid':
-            continue
-
-        all_results[algo] = load_one_experiment(system, task, algo, mpsc_cost_horizon)
-
-    return all_results
-
-
 def load_all_models(system, task, algo):
     '''Loads the results of every MPSC cost function for a specific experiment with every algo.
 
@@ -566,7 +519,7 @@ def plot_model_comparisons(system, task, algo, data_extractor):
     ax.set_ylabel(ylabel, weight='bold', fontsize=45, labelpad=10)
 
     x = np.arange(1, len(labels) + 1)
-    ax.set_xticks(x, labels, weight='bold', fontsize=15, rotation=45, ha='right')
+    ax.set_xticks(x, labels, weight='bold', fontsize=15, rotation=30, ha='right')
 
     medianprops = dict(linestyle='--', linewidth=2.5, color='black')
     bplot = ax.boxplot(data, patch_artist=True, labels=labels, medianprops=medianprops, widths=[0.75] * len(labels))
@@ -629,9 +582,6 @@ def plot_all_logs(system, task, algo):
         for seed in os.listdir(f'./models/rl_models/{system}/{task}/{algo}/{model}/'):
             all_results[model].append(load_from_logs(f'./models/rl_models/{system}/{task}/{algo}/{model}/{seed}/logs/'))
 
-    # all_results['safe_ppo'] = load_from_logs(f'./models/rl_models/{system}/{task}/safe_explorer_ppo/none/logs/')
-    # all_results['cpo'] = load_from_logs(f'./models/rl_models/{system}/{task}/cpo/none/logs/')
-
     for key in all_results['none'][0].keys():
         plot_log(system, task, algo, key, all_results)
 
@@ -655,12 +605,13 @@ def plot_log(system, task, algo, key, all_results):
     colors = {'mpsf_sr_pen_1': 'lightgreen', 'mpsf_sr_pen_10': 'limegreen', 'mpsf_sr_pen_100': 'forestgreen', 'mpsf_sr_pen_1000': 'darkgreen', 'none': 'cornflowerblue', 'none_cpen': 'plum'}
 
     for model in labels:
-        x = all_results[model][0][key][1]
+        x = all_results[model][0][key][1] / 1000
         all_data = np.array([values[key][3] for values in all_results[model]])
         ax.plot(x, np.mean(all_data, axis=0), label=model, color=colors[model])
         ax.fill_between(x, np.min(all_data, axis=0), np.max(all_data, axis=0), alpha=0.3, edgecolor=colors[model], facecolor=colors[model])
 
     ax.set_ylabel(key, weight='bold', fontsize=45, labelpad=10)
+    ax.set_xlabel('Training Episodes')
     ax.legend()
 
     fig.tight_layout()

experiments/mpsc/train_rl.py

@@ -1,9 +1,9 @@
 '''Template training/plotting/testing script.'''
 
-from functools import partial
 import os
 import shutil
 import time
+from functools import partial
 
 import munch
 import yaml

safe_control_gym/controllers/cpo/cpo.py

@@ -499,7 +499,7 @@ def get_cost(self, info):
         state_constraints = np.maximum(info['constraint_values'][:nx], info['constraint_values'][nx:nx * 2])
         constraint_width = info['constraint_values'][:nx] + info['constraint_values'][nx:nx * 2]
         state_cost = np.divide(state_constraints, -constraint_width / 2) + 0.0001
-        return np.sum([max(s,0) for s in state_cost])
+        return np.sum([max(s, 0) for s in state_cost])
 
     def normalizeAction(self, a):
         return normalize(a, self.action_bound_max, self.action_bound_min)

safe_control_gym/utils/logging.py

@@ -2,8 +2,8 @@
 
 import logging
 import os
-from collections import defaultdict
 import time
+from collections import defaultdict
 
 import imageio
 import numpy as np