Skip to content

Commit

Permalink
plot.py real robot plot, plus workaround for depth camera noise
Browse files Browse the repository at this point in the history
  • Loading branch information
@ahundt
ahundt committed Feb 9, 2020
1 parent a2a8aea commit 2d09f0d
Showing 1 changed file with 42 additions and 19 deletions.
61 changes: 42 additions & 19 deletions plot.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -190,7 +190,21 @@ def get_grasp_action_efficiency(actions, rewards, reward_threshold=0.5, window=2
best_success_rate(efficiency, window, 'grasp action efficiency')
return efficiency, lower, upper

def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:orange', 'tab:purple'], alpha=0.35, mult=100, max_iter=None, place=None, rasterized=True):

def real_robot_speckle_noise_hotfix(heights, trial, trial_success, clearance, over_height_threshold=6.0):
# length = min([heights.shape[0], trial.shape[0], trial_success.shape[0]])
actions_with_height_noise = heights > over_height_threshold
new_clearance = []
for trial_it in clearance:
recent_actions = actions_with_height_noise[int(trial_it) - 3:int(trial_it)]
if not np.any(recent_actions):
new_clearance += [trial_it]
trial = np.array(utils.clearance_log_to_trial_count(new_clearance)).astype(np.int)
heights[actions_with_height_noise] = 1.0
return heights, trial, trial_success, clearance


def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:orange', 'tab:purple'], alpha=0.35, mult=100, max_iter=None, place=None, rasterized=True, clear_figure=True, apply_real_robot_speckle_noise_hotfix=False):
stack_height_file = os.path.join(log_dir, 'transitions', 'stack-height.log.txt')
if os.path.isfile(stack_height_file):
heights = np.loadtxt(stack_height_file)
Expand Down Expand Up @@ -220,6 +234,25 @@ def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:o
# old versions of logged code don't have the grasp-success.log.txt file, data must be extracted from rewards.
grasp_rewards = rewards

# create and clear the figure
fig = plt.figure()
if clear_figure:
fig.clf()
# Plot the rate and variance of trial successes
trial_success_file = os.path.join(log_dir, 'transitions', 'trial-success.log.txt')
if os.path.isfile(trial_success_file):
trial_successes = np.loadtxt(trial_success_file)
if max_iter is not None:
trial_successes = trial_successes[:max_iter]
if apply_real_robot_speckle_noise_hotfix:
clearance = np.loadtxt(os.path.join(log_dir, 'transitions', 'clearance.log.txt'))
heights, trials, trial_successes, clearance = real_robot_speckle_noise_hotfix(heights, trials, trial_successes, clearance)
if trial_successes.size > 0:
trial_success_rate, trial_success_lower, trial_success_upper = get_trial_success_rate(trials, trial_successes, window=window)
plt.plot(mult*trial_success_rate, color=colors[3], label='Trial Success Rate')
plt.fill_between(np.arange(1, trial_success_rate.shape[0]+1),
mult*trial_success_lower, mult*trial_success_upper,
color=colors[3], alpha=alpha)
# trial_reward_file = os.path.join(log_dir, 'transitions', 'trial-reward-value.log.txt')
# if os.path.isfile(trial_reward_file):
# grasp_rewards = np.loadtxt(trial_reward_file)
Expand All @@ -234,9 +267,6 @@ def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:o
else:
eff, eff_lower, eff_upper = get_grasp_action_efficiency(actions, grasp_rewards, window=window)

# create and clear the figure
fig = plt.figure()
fig.clf()
plt.plot(mult*grasp_rate, color=colors[0], label='Grasp Success Rate')
if place:
plt.plot(mult*place_rate, color=colors[1], label='Place Success Rate')
Expand All @@ -253,19 +283,6 @@ def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:o
mult*eff_lower, mult*eff_upper,
color=colors[2], alpha=alpha)

# Plot the rate and variance of trial successes
trial_success_file = os.path.join(log_dir, 'transitions', 'trial-success.log.txt')
if os.path.isfile(trial_success_file):
trial_successes = np.loadtxt(trial_success_file)
if max_iter is not None:
trial_successes = trial_successes[:max_iter]
if trial_successes.size > 0:
trial_success_rate, trial_success_lower, trial_success_upper = get_trial_success_rate(trials, trial_successes, window=window)
plt.plot(mult*trial_success_rate, color=colors[3], label='Trial Success Rate')
plt.fill_between(np.arange(1, trial_success_rate.shape[0]+1),
mult*trial_success_lower, mult*trial_success_upper,
color=colors[3], alpha=alpha)

ax = plt.gca()
plt.xlabel('Number of Actions')
plt.ylabel('Mean % Over ' + str(window) + ' Actions, Higher is Better')
Expand All @@ -287,7 +304,13 @@ def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:o
log_dir = './logs/2020-01-20-11-40-56_Sim-Push-and-Grasp-Trial-Reward-Training'
log_dir = './logs/2020-01-20-14-25-13_Sim-Push-and-Grasp-Trial-Reward-Training'
log_dir = './logs/2020-02-03-14-47-16_Sim-Stack-Trial-Reward-Common-Sense-Training'

#############################################################
# REAL ROBOT STACKING run
plot_it('./logs/2020-02-09-11-02-57_Real-Stack-SPOT-Trial-Reward-Common-Sense-Training','Real Stack, SPOT Reward, Common Sense, Training', window=200, max_iter=None, apply_real_robot_speckle_noise_hotfix=True)
# Max trial success rate: 0.5833333333333334, at action iteration: 449. (total of 737 actions, max excludes first 200 actions)
# Max grasp success rate: 0.794392523364486, at action iteration: 289. (total of 750 actions, max excludes first 200 actions)
# Max place success rate: 0.7582417582417582, at action iteration: 119. (total of 751 actions, max excludes first 200 actions)
# Max action efficiency: 0.3, at action iteration: 37. (total of 751 actions, max excludes first 200 actions)
#############################################################
# Here is the good & clean simulation common sense push & grasp densenet plot with SPOT reward, run on the costar workstation.
# It can basically complete trials 100% of the time within 400 actions!
Expand All @@ -297,7 +320,7 @@ def plot_it(log_dir, title, window=1000, colors=['tab:blue', 'tab:green', 'tab:o
# ABSOLUTE BEST STACKING RUN AS OF 2020-02-04, on costar workstation
log_dir = './logs/2020-02-03-16-57-28_Sim-Stack-Trial-Reward-Common-Sense-Training'
# plot_it(log_dir, 'Sim Stack, Trial Reward, Common Sense, Training', window=window, max_iter=max_iter)
plot_it(log_dir,'Sim Stack, SPOT Reward, Common Sense, Training', window=window, max_iter=max_iter)
plot_it(log_dir,'Sim Stack, SPOT Reward, Common Sense, Training', window=window, max_iter=4000)
#############################################################

log_dir = './logs/2020-01-22-19-10-50_Sim-Push-and-Grasp-Two-Step-Reward-Training'
Expand Down

0 comments on commit 2d09f0d

Please sign in to comment.