Added sample of limitaiton

sample/analytical/limitation/best_result.txt

@@ -0,0 +1,7 @@
+nprocs = 10
+rank = 2
+step = 4523
+walker = 0
+fx = 0.00010188398524402734
+x1 = 3.584944906595298
+x2 = -1.8506985826548874

sample/analytical/limitation/do.sh

@@ -0,0 +1,3 @@
+mpiexec -np 10 --oversubscribe python3 ../../../src/py2dmat_main.py input.toml
+
+python3 hist2d_limitation_sample.py -p 10 -i input.toml -b 0.1

sample/analytical/limitation/hist2d_limitation_sample.py

@@ -0,0 +1,210 @@
+import numpy as np
+import matplotlib.pyplot as plt
+import matplotlib.colors as clr
+import toml as tm
+import sys
+import datetime
+import os
+import math as m
+import collections as cl
+import argparse
+import decimal
+
+def read_toml():
+    #input.tomlを取得
+    dict_toml= tm.load(inputfile)
+
+    #変数の範囲の最大値・最小値とTがlogスケールかを取得
+    x1_min,x2_min= dict_toml['algorithm']['param']['min_list']
+    x1_max,x2_max= dict_toml['algorithm']['param']['max_list']
+    T_min= dict_toml['algorithm']['exchange']['Tmin']
+    T_max= dict_toml['algorithm']['exchange']['Tmax']
+    Tlog= True #dict_toml['algorithm']['exchange']['Tlogspace']
+    #Tlogはブーリアン型
+
+    #結果を出力したディレクトリ名を取得
+    output_dicname= dict_toml['base']['output_dir']
+
+    #探索回数を取得
+    numsteps= int(dict_toml['algorithm']['exchange']['numsteps'])
+    numsteps_exchange= int(dict_toml['algorithm']['exchange']['numsteps_exchange'])
+
+    name_for_simulate = ''
+    return x1_min, x2_min, x1_max, x2_max, T_max, T_min, Tlog, output_dicname, numsteps, numsteps_exchange, name_for_simulate
+
+
+def read_result():
+
+    #各ディレクトリから得た値を格納する変数
+    x1= []
+    x2= []
+    fx= []
+    T= []
+    step= range(1,numsteps+1,1)
+
+    #各探索プロセッサにおけるfxの最大値・最小値を格納する変数
+    fx_max_list= []
+    fx_min_list= []
+
+    #以下よりファイルの読み込み
+    i= 0
+    while i <= number_of_replica-1 :
+        print('{0}/{1} ファイル読み込み開始'.format(i+1,number_of_replica))
+        #リストの中にi番目のディレクトリの値を格納するリストを作成
+        x1.append([])
+        x2.append([])
+        fx.append([])
+        T.append([])
+
+        #ファイルを開き、改行区切りのリストにする。
+        f=open('{0}/{1}/result.txt'.format(output_dicname,i))
+        frs= f.read().splitlines()
+        f.close()
+
+        #繰り返し処理で値を抽出、float型へ変換
+        #jが1からスタートなのはresult.txtの1行目には結果の値が入っていないため
+        j= 1
+        while j< len(frs):
+            sp= frs[j].split()
+            x1[i].append(float(sp[-2]))
+            x2[i].append(float(sp[-1]))
+            fx[i].append(float(sp[-3]))
+            T[i].append(float(sp[-4]))
+            j+= 1
+
+        #i番目のresult.txtにおけるfxの最大値・最小値を格納
+        fx_max_list.append(max(fx[i]))
+        fx_min_list.append(min(fx[i]))
+        i+= 1
+
+    print('読み込み完了')
+
+    #読み込みができてるかの確認
+    #print(x1[0][0:5],'\n\n',x1[1][0:5],'\n\n',x1[2][0:5],'\n\n',x1[3][0:5])
+    return x1, x2, fx, T, step, max(fx_max_list), min(fx_min_list)   
+
+
+def sort_T():
+    sorted_x1= []
+    sorted_x2= []
+    sorted_fx= []
+    replica_num= []
+    #original_x1= cp.deepcopy(x1)
+    #original_x2= cp.deepcopy(x2)
+    #original_fx= cp.deepcopy(fx)
+
+    Tvalue= make_Tvalue_list()
+
+    i= 0
+    while i < len(Tvalue):
+        sorted_x1.append([])
+        sorted_x2.append([])
+        sorted_fx.append([])
+        replica_num.append([])
+        i+= 1
+
+    i = int(numsteps*burn_in)
+    while i < numsteps:
+        j= 0
+        while j < number_of_replica:
+            k= 0
+            while k < len(Tvalue):
+                if Tvalue[k] == T[j][i]:
+                    sorted_x1[k].append(x1[j][i])   
+                    sorted_x2[k].append(x2[j][i])
+                    sorted_fx[k].append(fx[j][i])
+                    replica_num[k].append(j)
+                k+= 1
+            j+= 1
+        i+= 1
+
+    sorted_T= True
+    return sorted_x1, sorted_x2, sorted_fx, replica_num, Tvalue, sorted_T
+
+
+def make_Tvalue_list():
+    Tvalue= []
+    i= 0
+    while i < number_of_replica :
+        Tvalue.extend(list(cl.Counter(T[i]).keys()))
+        i+= 1
+    Tvalue= list(cl.Counter(Tvalue).keys())
+    return Tvalue
+
+def soloplot():
+    #繰り返し処理でプロット・図の保存
+    print('グラフ描画開始')
+    min_l = []
+    max_l = []
+
+    l= 0
+    while l< number_of_replica:
+        fig = plt.figure(figsize= (8,8))
+        ax = fig.add_subplot()
+
+        num_of_sample = len(sorted_x1[l][:])
+        weight_l = np.ones(num_of_sample)/num_of_sample
+
+        hst2d = ax.hist2d(
+                sorted_x1[l][:], sorted_x2[l][:],
+                norm= clr.LogNorm(vmin=10**-4, vmax=10**-1),
+                range=[ [x1_min,x1_max] , [x2_min,x2_max] ],
+                cmap= 'Reds', weights=weight_l , bins=100)
+
+        sum_allspace = np.sum(hst2d[0])
+        max_in_one_space = hst2d[0].max()
+        min_in_one_space = hst2d[0][np.nonzero(hst2d[0])].min()
+        min_l.append(min_in_one_space)
+        max_l.append(max_in_one_space)
+
+        cb = fig.colorbar(hst2d[3])
+
+        temp_for_title = decimal.Decimal(Tvalue[l]).quantize(
+                decimal.Decimal('0.001'), rounding=decimal.ROUND_HALF_UP
+                    )
+        figtitlestr = f'τ = {temp_for_title}'
+
+        ax.set_title(figtitlestr)
+
+        line_x1 = np.arange(x1_min,x1_max+1,1)
+        line_x2_1 = line_x1
+        line_x2_2 = -line_x1 + 1
+        ax.plot(line_x1,line_x2_1,c="black",alpha=0.3,lw=0.5)
+        ax.plot(line_x1,line_x2_2,c="black",alpha=0.3,lw=0.5)
+        #'''
+        ax.set_xlabel("x1")
+        ax.set_ylabel("x2")
+        ax.set_xlim(x1_min,x1_max)
+        ax.set_ylim(x2_min,x2_max)
+        ax.set_aspect('equal')
+        #軸サイズ設定と図の保存
+        fig.savefig(f'{dirname}/{l}_T_{temp_for_title}_burn_in_{burn_in}.png',dpi=300)             
+        l+= 1
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('-p','--mpiprocess',help='MPI process')
+    parser.add_argument('-i','--inputfile',help='input toml file')
+    parser.add_argument('-b','--burn_in',help='burn-in ratio')
+    args = parser.parse_args()
+
+    number_of_replica = int(args.mpiprocess)
+    inputfile = args.inputfile
+    burn_in = float(args.burn_in)
+
+    x1_min, x2_min, x1_max, x2_max, T_max, T_min, Tlog, output_dicname, numsteps, numsteps_exchange, name_for_simulate= read_toml()
+
+    #result.txtから読み取り
+    x1, x2, fx, T, step, fx_max, fx_min = read_result()
+
+    #グラフタイトル・ファイル名用に実行した時間の年日時間分を取得
+    time= datetime.date.today().strftime("%Y%m%d")
+
+    sorted_x1, sorted_x2, sorted_fx, replica_num, Tvalue, sorted_T= sort_T()
+
+    dirname= '{0}_histogram'.format(time)
+    if os.path.exists(dirname) == False:
+        os.makedirs(dirname)
+
+    soloplot()

sample/analytical/limitation/input.toml

@@ -0,0 +1,27 @@
+[base]
+dimension = 2
+output_dir = "output"
+
+[algorithm]
+name = "exchange"
+seed = 12345
+
+[algorithm.param]
+max_list = [6.0, 6.0]
+min_list = [-6.0, -6.0]
+unit_list = [0.3, 0.3]
+
+[algorithm.exchange]
+Tmin = 1.0
+Tmax = 100000.0
+numsteps = 10000
+numsteps_exchange = 100
+
+[solver]
+name = "analytical"
+function_name = "himmelblau"
+
+[runner]
+[runner.limitation]
+co_a = [[1, -1],[1, 1]]
+co_b = [[0, -1]]