-
Notifications
You must be signed in to change notification settings - Fork 2
/
Copy pathread_paper_tests.py
178 lines (154 loc) · 6.26 KB
/
read_paper_tests.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
import os
import pickle
import sys
import numpy as np
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
def read_experiment_observables(root_dir, key='tests_all'):
data_all = pd.DataFrame()
n_average = 5
for root, dirs, files in os.walk(root_dir):
file_names = []
for file in files:
if 'observables' in file and not 'Noise-off' in root:
file_names.append(file)
if file_names:
file_names.sort()
file_name = file_names[-1]
file_name = root + os.sep + file_name
openfile = open(file_name, 'rb')
if not key in ['step_counts_all']:
object = pickle.load(openfile)
data = pd.DataFrame(object[key]).sum()
# index = object['step_counts_all']
# data.index = index
data_frame = pd.DataFrame(data, columns=[root.split('_')[1]])
data_frame = data_frame.rolling(n_average).mean().shift(int(n_average / 2))
else:
data = pickle.load(openfile)[key]
print(root.split('_')[1])
data_frame = pd.DataFrame(data, columns=[root.split('_')[1]])
data_all = pd.concat([data_all, data_frame.T], sort=False)
openfile.close()
return data_all
def read_experiment_observables(root_dir, key='tests_all'):
data_all = pd.DataFrame()
n_average = 5
for root, dirs, files in os.walk(root_dir):
file_names = []
for file in files:
if 'observables' in file and not 'Noise-off' in root:
file_names.append(file)
if file_names:
file_names.sort()
file_name = file_names[-1]
file_name = root + os.sep + file_name
openfile = open(file_name, 'rb')
if not key in ['step_counts_all'] and not root.split('_')[1] == 'Five':
object = pickle.load(openfile)
data = pd.DataFrame(object[key][0], columns=[root.split('_')[1]])
index = object['step_counts_all']
data.index = index
data = data.groupby(data.index).max()
# print(data)
# else:
# data = pickle.load(openfile)[key]
# print(root.split('_')[1])
# data_frame = pd.DataFrame(data, columns=[root.split('_')[1]])
data_all = pd.concat([data_all, data], axis=1, sort=False)
openfile.close()
return data_all
def read_rewards(rewards):
stds = []
iterations = []
final_rews = []
mean_rews = []
for i in range(len(rewards)):
if len(rewards[i]) > 0:
final_rews.append(rewards[i][len(rewards[i]) - 1])
iterations.append(len(rewards[i]))
try:
mean_rews.append(np.sum(rewards[i][1:]))
except:
mean_rews.append([])
stds.append(np.std(rewards[i][1:]))
return np.array(iterations), np.array(final_rews), np.array(mean_rews), np.array(stds)
def read_experiment_final(root_dir):
data_all = pd.DataFrame()
n_average = 5
for root, dirs, files in os.walk(root_dir):
file_names = []
for file in files:
if 'final' in file and not 'Noise-off' in root:
file_names.append(file)
if file_names:
file_names.sort()
file_name = file_names[-1]
file_name = root + os.sep + file_name
print(file_name)
openfile = open(file_name, 'rb')
# data = np.sum(pickle.load(openfile)['rews'], axis=1)
data = pd.DataFrame(pickle.load(openfile)['rews']).sum()
# data_frame = pd.DataFrame(data, columns=[root.split('/')[0]])
data_frame = pd.DataFrame(data, columns=[root.split('_')[4]])
data_frame = data_frame.rolling(n_average).mean().shift(int(n_average / 2))
data_all = pd.concat([data_all, data_frame.T], sort=False)
openfile.close()
return data_all
# root_dir = 'Data_logging/Experiment_noise_pendulum/'
#
# figure_name = '../../Figures/Comparison_noise_ae_dyna.pdf'
#
# fig, axs = plt.subplots(3, 1, sharex=True)
# ax=axs[0]
# data_all_observables = read_experiment_observables(root_dir=root_dir, key='tests_all')
# data_all_observables = data_all_observables.loc[:, :60]
# sns.set_palette(sns.color_palette("bright", 3))
# sns.lineplot(ax=ax, data=data_all_observables.T, legend=False)
# # ax.set_title('Verification on real environment')
# ax.axhline(y=-200, ls=':', color = 'lime')
# ax.set_ylabel('mean cum. reward\n (arb. units.)')
#
# ax = axs[1]
# data_all_observables = read_experiment_observables(root_dir=root_dir, key='step_counts_all')
# data_all_observables = data_all_observables.loc[:, :60]
# sns.set_palette(sns.color_palette("bright", 8))
# sns.lineplot(ax=ax, data=data_all_observables.T)
# ax.set_ylabel('no. data points')
#
# ax=axs[2]
# data_all_observables = read_experiment_observables(root_dir=root_dir, key='batch_rews_all')
# data_all_observables = data_all_observables.loc[:, :60]
# sns.set_palette(sns.color_palette("bright", 8))
# sns.lineplot(ax=ax, data=data_all_observables.T, legend=False)
# ax.set_ylabel('batch reward\n (arb. units.)')
# ax.set_xlabel('no. epochs')
# # ax.set_title('Batch rewards during training')
#
# plt.tight_layout()
#
# plt.savefig(root_dir + figure_name)
# plt.show()
root_dir = 'Data/Simulation/ModelSizeLong/'
n_average = 30
figure_name = 'Figures/Compare_models_sizes.pdf'
data_all = read_experiment_observables(root_dir=root_dir)
data_all = data_all.reindex(np.linspace(201, 2010, num=151)).interpolate('values')
data_all = data_all.rolling(n_average).mean().shift(int(n_average / 2)).T
print(data_all.T)
fig, axs = plt.subplots(1, 1)
sns.set_palette(sns.color_palette("bright", 8))
sns.lineplot(ax=axs, data=data_all.T)
plt.ylabel('mean max cum. reward (arb. units.)')
plt.xlabel('no. data points')
plt.axhline(y=-200, ls=':', color='lime')
# plt.xlim(200,1600)
plt.savefig(figure_name)
plt.show()
data_all_rew = read_experiment_observables(root_dir=root_dir)
# print(data_all_rew)
# data_all_nr = read_experiment_observables(root_dir=root_dir, key='step_counts_all')
# print(data_all_nr)
# data_all = pd.concat([data_all_rew, data_all_nr], keys=['rew', 'nr'])
# print(data_all.T)