update new exp
This commit is contained in:
Caiyishuai
parent
41fbf7fb84
commit
fc2f1375f4
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,10,10.0,0,995.443,475.506,509.51,OBTEA,12.793,13.714,26.143,27.594,33.139,41.268,23.79,20.281,58.529,51.765,1.537,0.966,4.861,4.546,4.86,4.55,0.0017,0.002,3.39875
|
||||
1,10,100,10,10.0,0,995.443,475.506,509.51,Baseline,20.696,17.992,27.41,35.819,0.0,0.0,31.709,43.635,71.257,57.406,1.416,0.927,6.972,5.332,1.86,2.11,0.0011,0.00149,2.19649
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,10,10.0,5,995.058,466.107,527.256,OBTEA,23.272,60.593,34.341,68.761,54.706,146.318,28.532,45.897,26.114,30.608,1.545,1.09,8.384,20.177,8.38,20.18,0.00357,0.00914,7.14165
|
||||
1,10,100,10,10.0,5,995.058,466.107,527.256,Baseline,67.184,69.378,43.039,122.137,0.0,0.0,54.019,183.587,71.657,62.511,1.425,1.026,22.449,22.362,1.93,2.45,0.00198,0.00373,3.96889
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,10,50.0,0,1000.0,4046.512,2495.861,OBTEA,16.111,22.703,25.573,30.389,33.828,42.204,22.724,18.911,26.759,24.082,1.356,0.837,6.015,7.582,6.02,7.58,0.00956,0.0138,19.1263
|
||||
1,10,100,10,50.0,0,1000.0,4046.512,2495.861,Baseline,33.241,25.208,20.27,22.877,0.0,0.0,25.723,30.478,51.987,40.115,1.043,0.531,11.406,8.101,1.13,0.85,0.00272,0.00246,5.43922
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,30,10.0,5,2984.408,501.032,530.427,OBTEA,119.633,260.732,254.293,639.648,1005.143,2675.198,358.191,729.264,81.124,69.029,3.376,2.191,40.339,86.835,40.34,86.83,0.03694,0.09503,73.87664
|
||||
1,10,100,30,10.0,5,2984.408,501.032,530.427,Baseline,136.216,150.261,194.605,487.731,0.0,0.0,597.145,1580.933,170.94,124.32,3.335,2.211,43.98,48.066,6.28,6.79,0.01011,0.01305,20.21456
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,50,10.0,5,4974.567,497.667,524.377,OBTEA,215.587,491.812,524.244,1074.8,3578.785,11082.59,1210.47,2839.249,124.774,89.377,4.905,2.725,72.283,163.83,72.28,163.83,0.09599,0.26603,191.97842
|
||||
1,10,100,50,10.0,5,4974.567,497.667,524.377,Baseline,168.637,186.835,311.551,947.616,0.0,0.0,1581.167,7435.79,241.424,148.259,4.885,2.752,53.841,59.581,9.12,8.83,0.01843,0.02277,36.86494
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,50,30.0,5,4999.999,2495.997,1527.44,OBTEA,212.917,450.63,471.399,1225.393,3111.278,8152.979,1147.938,3147.285,97.673,79.408,3.973,2.446,71.409,150.106,71.41,150.11,0.25034,0.56939,500.6753
|
||||
1,10,100,50,30.0,5,4999.999,2495.997,1527.44,Baseline,202.099,240.551,288.256,661.414,0.0,0.0,1391.884,3429.381,189.436,136.676,3.856,2.464,64.124,75.363,11.73,16.21,0.05361,0.07497,107.22063
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,100,50,50.0,5,5000.0,4501.615,2531.486,OBTEA,253.295,572.722,409.349,916.217,2848.044,6624.733,971.793,2263.808,85.078,68.663,3.476,2.107,84.907,190.78,84.91,190.78,0.51015,1.2373,1020.29138
|
||||
1,10,100,50,50.0,5,5000.0,4501.615,2531.486,Baseline,254.321,442.356,318.881,994.072,0.0,0.0,1584.397,5523.477,169.383,124.202,3.311,2.154,80.446,137.838,14.99,31.82,0.11518,0.26258,230.35046
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,300,50,50.0,5,15000.0,13550.712,7560.888,OBTEA,220.524,466.348,382.831,847.677,2577.159,6096.284,892.785,2057.554,82.546,67.431,3.516,2.101,73.96,155.348,73.96,155.35,1.43287,3.01242,2865.73338
|
||||
1,10,300,50,50.0,5,15000.0,13550.712,7560.888,Baseline,242.892,346.826,276.286,627.216,0.0,0.0,1302.383,3431.004,164.248,115.986,3.342,2.122,77.08,108.217,13.66,24.46,0.29896,0.53628,597.9248
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,for_num_avg,for_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,500,50,50.0,0,25000.0,22468.606,12508.215,OBTEA,138.869,375.184,256.952,718.832,1779.607,6131.406,632.341,1979.175,137.008,105.888,3.375,2.009,46.787,124.961,46.79,124.96,1.53986,4.42153,3079.72499
|
||||
1,10,500,50,50.0,0,25000.0,22468.606,12508.215,Baseline,150.773,271.038,212.982,452.205,0.0,0.0,950.391,2086.918,164.186,116.935,3.376,2.147,45.514,78.291,16.24,37.74,0.57663,1.41181,1153.26951
|
||||
|
|
|
@ -0,0 +1,3 @@
|
|||
,depth,obj_num,cond_pred,act_pred,max_copy_times,literals_obj_count,state_avg,act_avg,btalgorithm,tree_size_avg,tree_size_std,ticks_avg,ticks_std,wm_cond_tick_avg,wm_cond_tick_std,cond_tick_avg,cond_tick_std,cost_avg,cost_std,step_avg,step_std,expand_num_avg,expand_num_std,plan_time_avg,plan_time_std,plan_time_total
|
||||
0,10,500,50,50.0,5,25000.0,22427.877,12514.209,OBTEA,226.104,482.321,393.348,915.23,2581.188,5635.137,909.768,2121.53,82.705,64.161,3.527,2.076,75.822,160.656,2.43283,5.09943,4865.66153
|
||||
1,10,500,50,50.0,5,25000.0,22427.877,12514.209,Baseline,250.694,351.381,306.653,847.463,0.0,0.0,1436.642,4042.081,165.935,119.651,3.321,2.106,79.464,109.247,0.51949,0.9819,1038.98435
|
||||
|
|
|
@ -240,43 +240,46 @@ np.random.seed(1)
|
|||
# #Condition Predicates
|
||||
# cond_pred_ls= 9* [10] + 9* [30] + 9*[50]
|
||||
|
||||
obj_num_ls = [10] #9* [100,300,500]
|
||||
# obj_num_ls = [10] #9* [100,300,500]
|
||||
# obj_num_ls = [100,100,500,\
|
||||
# 100,100,100,100,100,300,500]
|
||||
# Iterations Action Predicates
|
||||
act_pred_ls= [10]
|
||||
act_pred_ls = 3*act_pred_ls
|
||||
# cond_pred_ls = [10,10,50,\
|
||||
# 10,30,50,\
|
||||
# 50,50,50,50]
|
||||
# act_pred_ls= [10]
|
||||
# act_pred_ls = 3*act_pred_ls
|
||||
#Condition Predicates
|
||||
cond_pred_ls= [10]
|
||||
# act_pred_ls= [0,40,40,\
|
||||
# 0,0,0,\
|
||||
# 20,40,40,40]
|
||||
#
|
||||
#
|
||||
# max_copy_times_ls= [0,0] + 6*[5] #[5]*27
|
||||
|
||||
# max_copy_times_ls = [0]*3 + [5]*7
|
||||
# obj_num_ls = [300]
|
||||
# # Iterations Action Predicates
|
||||
# act_pred_ls= [40]
|
||||
# #Condition Predicates
|
||||
# cond_pred_ls= [50]
|
||||
|
||||
max_copy_times_ls= 27*[0]
|
||||
# max_copy_times_ls= 27*[0]
|
||||
|
||||
# obj_num_ls = [100]*5+[10,300,500]
|
||||
# cond_pred_ls=[10,30]+[50]*5
|
||||
# act_pred_ls=[0]*3+[20]+[40]*3
|
||||
# max_copy_times_ls=[5]*7
|
||||
#
|
||||
# depth_ls=[10]*27
|
||||
|
||||
|
||||
obj_num_ls = [300]
|
||||
cond_pred_ls=[50]
|
||||
act_pred_ls=[40]
|
||||
max_copy_times_ls=[5]*7
|
||||
|
||||
depth_ls=[10]*27
|
||||
|
||||
# obj_num_ls = [500]
|
||||
# # Iterations Action Predicates
|
||||
# act_pred_ls= [40]
|
||||
# #Condition Predicates
|
||||
# cond_pred_ls= [50]
|
||||
# max_copy_times_ls= [5]
|
||||
#
|
||||
# depth_ls=[10]*8
|
||||
|
||||
|
||||
# 文字数量 Cond Predicate
|
||||
# literals_num_ls= 5*[10]
|
||||
# literals_num_ls= [100,500,1000]
|
||||
# literals_num_ls= 4*[100]
|
||||
|
||||
# depth_ls= 100 * [10]
|
||||
|
||||
# max_copy_times_ls=[5,5]
|
||||
|
||||
|
||||
all_result=[]
|
||||
|
|
@ -302,8 +305,8 @@ for cond_pred,depth,act_pred,max_copy_times,obj_num in zip(cond_pred_ls, depth_l
|
|||
obt_result = BTTest_act_start_goal(bt_algo_opt=True, act_list=act_list, start_list=start_list, goal_list=goal_list,literals_num=literals_num)
|
||||
|
||||
|
||||
# baseline_result = BTTest_act_start_goal(bt_algo_opt=False, act_list=act_list, start_list=start_list,
|
||||
# goal_list=goal_list,literals_num=literals_num)
|
||||
baseline_result = BTTest_act_start_goal(bt_algo_opt=False, act_list=act_list, start_list=start_list,
|
||||
goal_list=goal_list,literals_num=literals_num)
|
||||
|
||||
|
||||
a_result=[]
|
||||
|
|
@ -319,28 +322,28 @@ for cond_pred,depth,act_pred,max_copy_times,obj_num in zip(cond_pred_ls, depth_l
|
|||
all_result.append(a_result)
|
||||
|
||||
|
||||
# import pandas as pd
|
||||
# df = pd.DataFrame(all_result, columns=[
|
||||
# # 'max_actcopy',
|
||||
# # 'literals_num','depth','iters','state','act',
|
||||
# # 'copy_act',
|
||||
# # 'depth', 'obj_num','iters','act_pred_num','literals_num','act_avg',
|
||||
# 'depth','obj_num', 'cond_pred', 'act_pred', 'max_copy_times', 'literals_obj_count', 'state_avg','act_avg',
|
||||
# 'btalgorithm',
|
||||
#
|
||||
# 'tree_size_avg', 'tree_size_std',
|
||||
# 'ticks_avg', 'ticks_std',
|
||||
# 'wm_cond_tick_avg','wm_cond_tick_std',
|
||||
# 'cond_tick_avg','cond_tick_std',
|
||||
# 'cost_avg', 'cost_std',
|
||||
# 'step_avg','step_std',
|
||||
# # 'state_num_avg','state_num_std',
|
||||
# 'expand_num_avg','expand_num_std',
|
||||
# 'for_num_avg','for_num_std',
|
||||
# 'plan_time_avg', 'plan_time_std', 'plan_time_total'])
|
||||
#
|
||||
# time_str = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime()).replace("-","").replace(":","")
|
||||
# csv_file_path = f'bt_randon_o={obj_num_ls[0]}_cp={cond_pred_ls[0]}_ap={act_pred_ls[0]+10}_MAE={max_copy_times_ls[0]}_time={time_str}.csv'
|
||||
# # param_ls = [depth, obj_num, cond_pred, act_pred_num, max_copy_times, literals_obj_count, state_avg, act_avg]
|
||||
# df.to_csv(csv_file_path, index=True)
|
||||
# print("CSV文件已生成:", csv_file_path)
|
||||
import pandas as pd
|
||||
df = pd.DataFrame(all_result, columns=[
|
||||
# 'max_actcopy',
|
||||
# 'literals_num','depth','iters','state','act',
|
||||
# 'copy_act',
|
||||
# 'depth', 'obj_num','iters','act_pred_num','literals_num','act_avg',
|
||||
'depth','obj_num', 'cond_pred', 'act_pred', 'max_copy_times', 'literals_obj_count', 'state_avg','act_avg',
|
||||
'btalgorithm',
|
||||
|
||||
'tree_size_avg', 'tree_size_std',
|
||||
'ticks_avg', 'ticks_std',
|
||||
'wm_cond_tick_avg','wm_cond_tick_std',
|
||||
'cond_tick_avg','cond_tick_std',
|
||||
'cost_avg', 'cost_std',
|
||||
'step_avg','step_std',
|
||||
# 'state_num_avg','state_num_std',
|
||||
'expand_num_avg','expand_num_std',
|
||||
'for_num_avg','for_num_std',
|
||||
'plan_time_avg', 'plan_time_std', 'plan_time_total'])
|
||||
|
||||
time_str = time.strftime('%Y-%m-%d %H:%M:%S', time.localtime()).replace("-","").replace(":","")
|
||||
csv_file_path = f'bt_randon_o={obj_num}_cp={cond_pred}_ap={act_pred+10}_MAE={max_copy_times}_time={time_str}.csv'
|
||||
# param_ls = [depth, obj_num, cond_pred, act_pred_num, max_copy_times, literals_obj_count, state_avg, act_avg]
|
||||
df.to_csv(csv_file_path, index=True)
|
||||
print("CSV文件已生成:", csv_file_path)
|
||||
|
|
@ -0,0 +1,72 @@
|
|||
import os
|
||||
import pandas as pd
|
||||
|
||||
# Directory where CSVs are located
|
||||
csv_dir = './'
|
||||
# List of CSV file names as you described
|
||||
csv_files = [
|
||||
'bt_randon_o=100_cp=10_ap=10_MAE=0_time=20240506 115257.csv',
|
||||
'bt_randon_o=100_cp=10_ap=10_MAE=5_time=20240506 172702.csv',
|
||||
'bt_randon_o=100_cp=10_ap=50_MAE=0_time=20240506 115600.csv',
|
||||
'bt_randon_o=100_cp=30_ap=10_MAE=5_time=20240508 222143.csv',
|
||||
'bt_randon_o=100_cp=50_ap=10_MAE=5_time=20240508 222622.csv',
|
||||
'bt_randon_o=100_cp=50_ap=30_MAE=5_time=20240508 223352.csv',
|
||||
'bt_randon_o=100_cp=50_ap=50_MAE=5_time=20240508 224801.csv',
|
||||
'bt_randon_o=300_cp=50_ap=50_MAE=5_time=20240508 232751.csv',
|
||||
'bt_randon_o=500_cp=50_ap=50_MAE=0_time=20240506 124308.csv',
|
||||
'bt_randon_o=500_cp=50_ap=50_MAE=5_time=20240120 020115.csv'
|
||||
]
|
||||
|
||||
|
||||
# Prepare a list to accumulate the results
|
||||
results = []
|
||||
# Loop through all CSV files
|
||||
for file in csv_files:
|
||||
# Extract the scenario details from the file name
|
||||
parts = file.split('_')
|
||||
scenario = {
|
||||
'Objects': int(parts[2].split('=')[1]),
|
||||
'Pc': int(parts[3].split('=')[1]),
|
||||
'Pa': int(parts[4].split('=')[1]),
|
||||
'MAE': int(parts[5].split('=')[1]),
|
||||
'Time': parts[6].split('=')[1] + '_' + parts[6].split('.')[0]
|
||||
}
|
||||
|
||||
# Load CSV into DataFrame
|
||||
df = pd.read_csv(os.path.join(csv_dir, file))
|
||||
|
||||
# Extract results for each algorithm
|
||||
obtea = df.iloc[0]
|
||||
baseline = df.iloc[1]
|
||||
|
||||
# Collect relevant data in a single row
|
||||
result = {
|
||||
'Objects': scenario['Objects'],
|
||||
'Pc': scenario['Pc'],
|
||||
'Pa': scenario['Pa'],
|
||||
'MAE': scenario['MAE'],
|
||||
'Literals': round(obtea['literals_obj_count'], 1),
|
||||
'States Avg': round(obtea['state_avg'], 1),
|
||||
'Actions Avg': round(obtea['act_avg'], 1),
|
||||
'Baseline Cost Avg': round(baseline['cost_avg'], 1),
|
||||
'OBTEA Cost Avg': round(obtea['cost_avg'], 1),
|
||||
'Baseline Cond Tick Avg': round(baseline['cond_tick_avg'], 1),
|
||||
'OBTEA Cond Tick Avg': round(obtea['cond_tick_avg'], 1),
|
||||
'OBTEA WM Cond Tick Avg': round(obtea['wm_cond_tick_avg'], 1)
|
||||
}
|
||||
# Add this result to the list
|
||||
results.append(result)
|
||||
|
||||
# Convert the list of results into a DataFrame
|
||||
final_results = pd.DataFrame(results)
|
||||
|
||||
# Sort the DataFrame
|
||||
final_results.sort_values(by=['MAE', 'Objects', 'Pc', 'Pa'], inplace=True)
|
||||
|
||||
# Display the final results
|
||||
# Output the entire DataFrame
|
||||
pd.set_option('display.max_columns', None)
|
||||
pd.set_option('display.width', None)
|
||||
print(final_results)
|
||||
# 输出的表格，每行的内容是 Objects、Pc、Pa、MAE，csv中的 literals_obj_count、state_avg、act_avg，
|
||||
# Baseline的cost_avg，OBTEA的cost_avg，Baseline的cond_tick_avg，OBTEA的cond_tick_avg，OBTEA的wm_cond_tick_avg
|
||||
Loading…
Reference in New Issue