遗传算法

.idea/csv-editor.xml

@@ -3,7 +3,42 @@
   <component name="CsvFileAttributes">
     <option name="attributeMap">
       <map>
-        <entry key="C:\Users\www\Desktop\python项目\数据\抽样第3次数据\firm_amended.csv">
+        <entry key="\GA_Agent_0925\convergence0119_data.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
+        <entry key="\GA_Agent_0925\count_firm.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
+        <entry key="\GA_Agent_0925\count_firm_prod.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
+        <entry key="\GA_Agent_0925\count_prod.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
+        <entry key="\GA_Agent_0925\risk_ay\count.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
+        <entry key="\GA_Agent_0925\risk_ay\count_dcp.csv">
           <value>
             <Attribute>
               <option name="separator" value="," />
@@ -24,6 +59,13 @@
             </Attribute>
           </value>
         </entry>
+        <entry key="\GA_Agent_0925\vulnerable35_match_results\vulnerable20_match_results_new260111.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
         <entry key="\GA_Agent_0925\vulnerable35_match_results\vulnerable35_match_results.csv">
           <value>
             <Attribute>
@@ -38,6 +80,13 @@
             </Attribute>
           </value>
         </entry>
+        <entry key="\GA_Agent_0925\vulnerable35_match_results\vulnerable63_match_results_new260111.csv">
+          <value>
+            <Attribute>
+              <option name="separator" value="," />
+            </Attribute>
+          </value>
+        </entry>
         <entry key="\count_firm.csv">
           <value>
             <Attribute>

.idea/dataSources.local.xml

@@ -1,6 +1,6 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <project version="4">
-  <component name="dataSourceStorageLocal" created-in="PY-242.23726.102">
+  <component name="dataSourceStorageLocal" created-in="PY-242.26775.22">
     <data-source name="@localhost" uuid="3ce7b935-0ff7-47a3-aaa8-91063c963644">
       <database-info product="MySQL" version="8.0.36" jdbc-version="4.2" driver-name="Amazon Web Services (AWS) Advanced JDBC Wrapper" driver-version="Amazon Web Services (AWS) Advanced JDBC Wrapper 2.3.7 ( Revision: 7591851e8da4e1c705ba232a8bd07824a5cfd276 )" dbms="MYSQL" exact-version="8.0.36" exact-driver-version="2.3">
         <extra-name-characters>#@</extra-name-characters>

.idea/dataSources/3ce7b935-0ff7-47a3-aaa8-91063c963644.xml

@@ -964,7 +964,7 @@ iiabmdb_20250926|schema||iiabm_user|localhost|UPDATE|G</Grants>
       <CollationName>utf8mb4_general_ci</CollationName>
     </schema>
     <schema id="289" parent="1" name="iiabmdb_20250925">
-      <LastIntrospectionLocalTimestamp>2025-11-21.16:00:59</LastIntrospectionLocalTimestamp>
+      <LastIntrospectionLocalTimestamp>2026-01-26.12:22:45</LastIntrospectionLocalTimestamp>
       <CollationName>utf8mb4_0900_ai_ci</CollationName>
     </schema>
     <schema id="290" parent="1" name="iiabmdb_20250926">
@@ -1311,7 +1311,7 @@ iiabmdb_20250926|schema||iiabm_user|localhost|UPDATE|G</Grants>
       <UnderlyingIndexName>PRIMARY</UnderlyingIndexName>
     </key>
     <column id="362" parent="299" name="id">
-      <AutoIncrement>11616939</AutoIncrement>
+      <AutoIncrement>22818</AutoIncrement>
       <NotNull>1</NotNull>
       <Position>1</Position>
       <StoredType>int|0s</StoredType>
@@ -1365,7 +1365,7 @@ iiabmdb_20250926|schema||iiabm_user|localhost|UPDATE|G</Grants>
       <UnderlyingIndexName>PRIMARY</UnderlyingIndexName>
     </key>
     <column id="373" parent="300" name="id">
-      <AutoIncrement>3631</AutoIncrement>
+      <AutoIncrement>364</AutoIncrement>
       <NotNull>1</NotNull>
       <Position>1</Position>
       <StoredType>int|0s</StoredType>

.idea/sqldialects.xml

@@ -2,6 +2,7 @@
 <project version="4">
   <component name="SqlDialectMappings">
     <file url="file://$PROJECT_DIR$/GA_Agent_0925/SQL_analysis_risk_ga.sql" dialect="MySQL" />
+    <file url="file://$PROJECT_DIR$/GA_Agent_0925/vulnerable35_match_results/query_top35_products.sql" dialect="MySQL" />
     <file url="file://$PROJECT_DIR$/SQL_analysis_experiment.sql" dialect="MySQL" />
     <file url="file://$PROJECT_DIR$/SQL_analysis_risk.sql" dialect="MySQL" />
     <file url="file://$PROJECT_DIR$/SQL_db_user_create.sql" dialect="MySQL" />

GA_Agent_0925/conf_experiment.yaml

@@ -8,5 +8,5 @@ test:  # only for test scenarios
   n_iter: 100
 
 not_test:  # normal scenarios
-  n_sample: 5
+  n_sample: 1
-  n_iter: 60
+  n_iter: 50

GA_Agent_0925/config.json

@@ -1,10 +1,10 @@
 {
-  "pop_size": 20,
+  "pop_size": 10,
-  "n_gen": 10,
+  "n_gen": 60,
   "n_var": 12,
   "bound_min": -5,
   "bound_max": 5,
-  "cx_prob": 0.5,
+  "cx_prob": 0.8,
   "mut_prob": 0.2,
   "cx_alpha": 0.5,
   "mut_sigma": 0.1,

GA_Agent_0925/controller_db.py

@@ -328,7 +328,7 @@ class ControllerDB:
                  )).scalar()
         # 控制 n_sample数量 作为后面的参数
         n_sample = 0 if res is None else res
-        # print(f'There are a total of {n_sample} samples.')
+        print(f'There are a total of {n_sample} samples.')
         # 查找 is_done_flag = -1 也就是没有运行的 样本 运行后会改为0
         res = db_session.execute(
             text(f"SELECT id FROM {self.db_name_prefix}_sample "

GA_Agent_0925/convergence0119_data.csv

@@ -0,0 +1,64 @@
+Timestamp,Generation,Best_Fitness_Percentage
+2026-01-26 21:42:26,0,0.2090216731702263
+2026-01-26 22:30:22,1,0.2052078225397421
+2026-01-26 21:42:26,2,0.2016374981412617
+2026-01-26 22:30:22,3,0.2013488674598191
+2026-01-26 21:58:42,4,0.1993414391454412
+2026-01-26 22:30:22,5,0.1988803893310773
+2026-01-26 21:10:30,6,0.1987272799730525
+2026-01-26 22:30:22,7,0.194483284141369
+2026-01-26 22:30:22,8,0.1940582914278802
+2026-01-26 21:42:26,9,0.1928494295868045
+2026-01-26 21:58:42,10,0.1898885073092019
+2026-01-26 21:42:26,11,0.1889512197824896
+2026-01-26 21:58:42,12,0.1858551470980505
+2026-01-26 21:42:26,13,0.1845113472201131
+2026-01-26 22:15:59,14,0.184417748143425
+2026-01-26 21:42:26,15,0.1826658943130832
+2026-01-26 21:42:26,16,0.1811738541290951
+2026-01-26 21:58:42,17,0.1803782808697044
+2026-01-26 21:58:42,18,0.180378251293931
+2026-01-26 21:10:30,19,0.1802064263206075
+2026-01-26 21:42:26,20,0.1799530185714123
+2026-01-26 21:42:26,21,0.1799530185714123
+2026-01-26 21:42:26,22,0.1799530185714123
+2026-01-26 21:58:42,23,0.1776493750092811
+2026-01-26 21:58:42,24,0.1776493750092811
+2026-01-26 21:58:42,25,0.1776493750092811
+2026-01-26 21:58:42,26,0.1776493750092811
+2026-01-26 21:58:42,27,0.1776493750092811
+2026-01-26 22:30:22,28,0.1774372678590745
+2026-01-26 22:30:22,29,0.1774372678590745
+2026-01-26 22:30:22,30,0.1774372678590745
+2026-01-26 22:30:22,31,0.1774372678590745
+2026-01-26 22:30:22,32,0.1774372678590745
+2026-01-26 22:30:22,33,0.1774372678590745
+2026-01-26 22:30:22,34,0.1774372678590745
+2026-01-26 22:30:22,35,0.1774372678590745
+2026-01-26 22:30:22,36,0.1774372678590745
+2026-01-26 22:30:22,37,0.1774372678590745
+2026-01-26 22:30:22,38,0.1774372678590745
+2026-01-26 22:30:22,39,0.1774372678590745
+2026-01-26 22:30:22,40,0.1774372678590745
+2026-01-26 22:30:22,41,0.1774372678590745
+2026-01-26 22:30:22,42,0.1774372678590745
+2026-01-26 22:30:22,43,0.1774372678590745
+2026-01-26 22:30:22,44,0.1774372678590745
+2026-01-26 22:30:22,45,0.1774372678590745
+2026-01-26 22:30:22,46,0.1774372678590745
+2026-01-26 22:30:22,47,0.1774372678590745
+2026-01-26 22:30:22,48,0.1774372678590745
+2026-01-26 22:30:22,49,0.1774372678590745
+2026-01-26 22:30:22,50,0.1774372678590745
+2026-01-26 22:30:22,51,0.1774372678590745
+2026-01-26 22:30:22,52,0.1774372678590745
+2026-01-26 22:30:22,53,0.1774372678590745
+2026-01-26 22:30:22,54,0.1774372678590745
+2026-01-26 22:30:22,55,0.1774372678590745
+2026-01-26 22:30:22,56,0.1774372678590745
+2026-01-26 22:30:22,57,0.1774372678590745
+2026-01-26 22:30:22,58,0.1774372678590745
+2026-01-26 22:30:22,59,0.1774372678590745
+2026-01-26 22:30:22,60,0.1774372678590745
+2026-01-26 22:30:22,61,0.1774372678590745
+2026-01-26 22:30:22,62,0.1774372678590745

GA_Agent_0925/count_firm.csv

@@ -0,0 +1,198 @@
+id_firm,count
+214851100,303
+3111603340,299
+70634828,282
+395736790,122
+340093034,117
+29223617,96
+25980377,81
+3330358736,79
+2337727838,63
+331545755,60
+2326722141,60
+3191869223,54
+517675473,53
+2327605629,51
+591350440,48
+728969035,46
+532328014,45
+2336923756,43
+303926772,38
+16116663,38
+471121089,37
+2349179532,35
+16210433,34
+2349705416,31
+1452048,5
+515770253,3
+2349345463,2
+6,2
+5971532,2
+1,2
+5849940,2
+2424229017,2
+495782506,2
+420984285,2
+300186799,2
+37873062,2
+3392803162,2
+33822284,2
+3312358902,2
+2728939,2
+3226664625,2
+79938367,2
+78979697,2
+14913649,2
+80158773,2
+8,2
+11807506,2
+169978927,2
+4607820,1
+441623911,1
+367669349,1
+410030851,1
+395739442,1
+808524154,1
+38852110,1
+385766513,1
+8114841,1
+382080545,1
+466148111,1
+805940123,1
+675729777,1
+863973253,1
+35404067,1
+3462551351,1
+3429928077,1
+3420061649,1
+340603317,1
+872394725,1
+9032550,1
+907433543,1
+950849442,1
+3378606529,1
+3362063909,1
+3358892171,1
+961017,1
+9620005,1
+862404568,1
+493002466,1
+668539285,1
+507827038,1
+676597455,1
+688155470,1
+664591135,1
+648145286,1
+644252759,1
+643954924,1
+640700057,1
+6333996,1
+618469306,1
+696450846,1
+71271700,1
+733657390,1
+3268669333,1
+5979030,1
+737770776,1
+742704658,1
+596368303,1
+59234665,1
+756272716,1
+758879940,1
+578803019,1
+562681526,1
+551856519,1
+543470507,1
+762501019,1
+771821595,1
+518871190,1
+778745779,1
+78576577,1
+598808584,1
+314846874,1
+3226232,1
+2333843479,1
+2310534839,1
+2311838590,1
+2314659369,1
+2316150629,1
+2323069589,1
+2326520912,1
+2326655246,1
+2326903290,1
+2334430421,1
+225958786,1
+2337952436,1
+2341774429,1
+23421122,1
+2344471631,1
+2345982379,1
+2347561020,1
+2348894245,1
+2348987001,1
+2310406050,1
+205960791,1
+3222821993,1
+1524794108,1
+1160497810,1
+1171244159,1
+1208566436,1
+1247902451,1
+1253552935,1
+13854344,1
+1444449910,1
+1476953321,1
+157087137,1
+203314437,1
+1605495,1
+1651310523,1
+16715045,1
+1698501971,1
+178452970,1
+189427260,1
+193814549,1
+197362120,1
+2349349655,1
+2349588257,1
+2349742676,1
+3145389278,1
+2978926070,1
+2989649772,1
+29954548,1
+3010580773,1
+3011933107,1
+3031766093,1
+3118140206,1
+3135349256,1
+3147958370,1
+2349746655,1
+10437056,1
+3151377261,1
+31732840,1
+3195293647,1
+3196033145,1
+3203980088,1
+3220049148,1
+3222664794,1
+2962064709,1
+29452962,1
+2944593082,1
+290636928,1
+2350443114,1
+2350544061,1
+2351592628,1
+2352421906,1
+2353389310,1
+2353549582,1
+2382390052,1
+2448521375,1
+247297633,1
+25685135,1
+26487185,1
+265133300,1
+26895145,1
+27042865,1
+280281699,1
+286335813,1
+28667694,1
+9746245,1

GA_Agent_0925/count_firm_prod.csv

@@ -0,0 +1,259 @@
+id_firm,id_product,count
+395736790,95,122
+340093034,95,117
+29223617,95,96
+2337727838,90,63
+2326722141,90,60
+331545755,90,60
+3191869223,91,54
+517675473,92,53
+2327605629,94,51
+591350440,91,48
+728969035,93,46
+532328014,99,45
+2336923756,91,43
+303926772,99,38
+471121089,93,37
+2349179532,93,35
+70634828,55,34
+16210433,92,34
+16116663,92,34
+214851100,52,33
+3111603340,52,33
+214851100,55,32
+70634828,52,31
+3111603340,53,31
+214851100,51,31
+3111603340,55,31
+2349705416,94,31
+3111603340,50,30
+70634828,53,30
+70634828,50,29
+214851100,53,27
+3111603340,54,27
+214851100,54,26
+214851100,50,26
+70634828,51,25
+3111603340,51,25
+214851100,49,24
+214851100,44,24
+3111603340,47,24
+214851100,48,23
+70634828,46,22
+214851100,47,22
+3111603340,48,21
+3111603340,46,20
+70634828,47,20
+3111603340,49,20
+70634828,54,20
+3330358736,41,19
+70634828,48,19
+3111603340,44,19
+70634828,49,18
+70634828,44,18
+3330358736,42,18
+25980377,41,18
+214851100,46,18
+3111603340,45,17
+214851100,45,16
+25980377,43,15
+70634828,45,15
+25980377,38,14
+25980377,42,13
+25980377,39,12
+3330358736,38,12
+3330358736,43,11
+3330358736,39,10
+3330358736,40,9
+25980377,40,9
+1452048,9,3
+515770253,9,3
+495782506,19,2
+420984285,16,2
+5971532,9,2
+37873062,9,2
+79938367,9,2
+8,37,2
+3392803162,9,2
+33822284,9,2
+80158773,69,2
+6,10,2
+1,10,2
+14913649,9,2
+16116663,11,2
+2350443114,74,1
+5979030,9,1
+562681526,25,1
+578803019,10,1
+5849940,26,1
+5849940,36,1
+16116663,9,1
+59234665,65,1
+596368303,12,1
+598808584,15,1
+543470507,8,1
+16116663,10,1
+618469306,9,1
+6333996,66,1
+640700057,9,1
+643954924,9,1
+644252759,61,1
+648145286,35,1
+664591135,9,1
+551856519,33,1
+1651310523,9,1
+675729777,10,1
+518871190,9,1
+35404067,9,1
+367669349,31,1
+193814549,33,1
+382080545,9,1
+385766513,24,1
+38852110,10,1
+189427260,9,1
+395739442,9,1
+410030851,27,1
+178452970,25,1
+441623911,79,1
+4607820,9,1
+466148111,9,1
+169978927,9,1
+493002466,9,1
+169978927,66,1
+507827038,10,1
+1698501971,9,1
+16715045,10,1
+668539285,17,1
+676597455,68,1
+2350544061,9,1
+808524154,9,1
+778745779,9,1
+78576577,9,1
+78979697,61,1
+78979697,74,1
+11807506,36,1
+1171244159,32,1
+1160497810,12,1
+805940123,72,1
+8114841,9,1
+762501019,9,1
+862404568,11,1
+863973253,9,1
+872394725,70,1
+9032550,34,1
+907433543,10,1
+950849442,9,1
+961017,23,1
+9620005,9,1
+771821595,31,1
+758879940,65,1
+688155470,30,1
+13854344,9,1
+696450846,9,1
+1605495,9,1
+157087137,73,1
+1524794108,9,1
+1476953321,10,1
+1452048,30,1
+1452048,11,1
+1444449910,20,1
+1253552935,9,1
+756272716,32,1
+1247902451,9,1
+1208566436,62,1
+70634828,7,1
+71271700,27,1
+11807506,9,1
+733657390,10,1
+737770776,34,1
+742704658,9,1
+3462551351,13,1
+3429928077,9,1
+3420061649,9,1
+340603317,9,1
+280281699,60,1
+286335813,72,1
+28667694,10,1
+290636928,9,1
+2348987001,9,1
+2944593082,9,1
+29452962,9,1
+2962064709,9,1
+2978926070,8,1
+2989649772,71,1
+29954548,18,1
+300186799,10,1
+300186799,11,1
+3010580773,9,1
+3011933107,9,1
+3031766093,9,1
+2348894245,11,1
+2347561020,9,1
+2345982379,67,1
+2728939,71,1
+2728939,63,1
+27042865,9,1
+247297633,18,1
+2351592628,10,1
+2352421906,17,1
+2353389310,15,1
+2353549582,23,1
+2382390052,9,1
+2424229017,26,1
+2424229017,9,1
+2448521375,25,1
+25685135,11,1
+26895145,9,1
+2349746655,10,1
+2349742676,33,1
+2349588257,10,1
+2349349655,9,1
+2349345463,59,1
+2349345463,35,1
+26487185,62,1
+265133300,9,1
+2344471631,9,1
+23421122,9,1
+2341774429,9,1
+2310534839,64,1
+3226664625,13,1
+3226664625,28,1
+3268669333,67,1
+3312358902,59,1
+3312358902,79,1
+2316150629,10,1
+2314659369,20,1
+2311838590,97,1
+2310406050,60,1
+3222821993,9,1
+225958786,11,1
+214851100,7,1
+3358892171,64,1
+3362063909,10,1
+3378606529,9,1
+205960791,63,1
+203314437,22,1
+197362120,15,1
+3226232,10,1
+3222664794,9,1
+2337952436,24,1
+3135349256,9,1
+2334430421,73,1
+2333843479,70,1
+10437056,9,1
+2326903290,9,1
+2326655246,9,1
+2326520912,10,1
+3111603340,7,1
+3118140206,68,1
+3145389278,9,1
+3220049148,9,1
+3147958370,12,1
+314846874,11,1
+3151377261,29,1
+31732840,29,1
+2323069589,9,1
+3195293647,28,1
+3196033145,9,1
+3203980088,10,1
+9746245,97,1

GA_Agent_0925/count_prod.csv

@@ -0,0 +1,73 @@
+id_product,count
+95,335
+90,183
+91,145
+92,121
+93,118
+55,97
+52,97
+53,88
+50,85
+9,84
+99,83
+94,82
+51,81
+54,73
+47,66
+48,63
+49,62
+44,61
+46,60
+45,48
+41,37
+42,31
+38,26
+43,26
+10,23
+39,22
+40,18
+11,9
+15,3
+33,3
+7,3
+12,3
+25,3
+18,2
+74,2
+68,2
+69,2
+70,2
+71,2
+72,2
+73,2
+97,2
+79,2
+8,2
+17,2
+19,2
+66,2
+13,2
+16,2
+67,2
+28,2
+65,2
+26,2
+30,2
+31,2
+32,2
+34,2
+35,2
+36,2
+37,2
+27,2
+24,2
+64,2
+23,2
+20,2
+29,2
+60,2
+61,2
+62,2
+63,2
+59,2
+22,1

GA_Agent_0925/creating.py

@@ -3,26 +3,29 @@ from deap import creator, base, tools
 from evaluate_func import fitness
 
 def creating():
-    if "FitnessMax" not in creator.__dict__:
+    if "FitnessMin" not in creator.__dict__:
-        creator.create("FitnessMax", base.Fitness, weights=(1.0,))
+        creator.create("FitnessMin", base.Fitness, weights=(-1.0,))
     if "Individual" not in creator.__dict__:
-        creator.create("Individual", list, fitness=creator.FitnessMax)
+        creator.create("Individual", list, fitness=creator.FitnessMin)
 
     toolbox = base.Toolbox()
 
-    # 基因注册
+    # ==============================
-    toolbox.register("n_max_trial", random.randint, 1, 60)
+    # 基因注册（修改后）
-    toolbox.register("prf_size", random.uniform, 0.0, 1.0)
+    # ==============================
-    toolbox.register("prf_conn", random.uniform, 0.0, 1.0)
+
+    toolbox.register("n_max_trial", random.randint, 50, 200)
+    toolbox.register("prf_size", random.uniform, 0.05, 0.95)
+    toolbox.register("prf_conn", random.uniform, 0.05, 0.95)
     toolbox.register("cap_limit_prob_type", random.randint, 0, 1)
-    toolbox.register("cap_limit_level", random.randint, 5, 80)
+    toolbox.register("cap_limit_level", random.randint, 10, 50)
-    toolbox.register("diff_new_conn", random.uniform, 0.0, 1.0)
+    toolbox.register("diff_new_conn", random.uniform, 0.05, 0.95)
-    toolbox.register("netw_prf_n", random.randint, 1, 20)
+    toolbox.register("netw_prf_n", random.randint, 3, 15)
-    toolbox.register("s_r", random.uniform, 0.01, 0.5)
+    toolbox.register("s_r", random.uniform, 0.05, 0.4)
-    toolbox.register("S_r", random.uniform, 0.5, 1.0)
+    toolbox.register("S_r", random.uniform, 0.8, 2.5)
-    toolbox.register("x", random.uniform, 0.0, 1)
+    toolbox.register("x", random.uniform, 0.1, 0.9)
-    toolbox.register("k", random.uniform, 0.1, 5.0)
+    toolbox.register("k", random.uniform, 0.2, 4.0)
-    toolbox.register("production_increase_ratio", random.uniform, 1, 5.0)
+    toolbox.register("production_increase_ratio", random.uniform, 0.3, 3.0)
 
     # 个体与种群注册
     toolbox.register(

GA_Agent_0925/evaluate_func.py

@@ -13,7 +13,6 @@ from my_model import MyModel
 from orm import connection, engine
 
 
-# 🎯 适应度函数（核心目标函数）
 def fitness(individual, controller_db_obj):
     """
     遗传算法适应度函数：用于评估个体（模型参数）的优劣。
@@ -57,7 +56,7 @@ def fitness(individual, controller_db_obj):
     #     print(f"  {key}: {value}")
     # ========== 2️⃣ 调用 ABM 模型 ==========
     # 并行进程数目
-    job=4
+    job=6
     do_process(controller_db_obj,ga_id,dct_exp,job)
     # ========== 3️⃣ 获取数据库连接并提取结果 ==========
     simulated_vulnerable_industries = get_vulnerable35_code(connection,ga_id)
@@ -65,25 +64,36 @@ def fitness(individual, controller_db_obj):
     # ========== 4️⃣ 获取目标产业集合 ==========
     target_vulnerable_industries = get_target_vulnerable_industries()
 
-    # ========== 5️⃣ 计算误差（集合差异度） ==========
+    """
-    set_sim = set(simulated_vulnerable_industries)
+        Top-K 加权命中误差（越小越好）
-    set_target = set(target_vulnerable_industries)
-    error = len(set_sim.symmetric_difference(set_target))
 
-    simulated_set = set(simulated_vulnerable_industries)
+        simulated_vulnerable_industries : list[str]
+            模型输出的产业排序（风险从高到低）
+        target_vulnerable_industries : list[str] or set[str]
+            真实脆弱产业集合（无序）
+        """
+    ranked_list = simulated_vulnerable_industries
     target_set = set(target_vulnerable_industries)
 
-    matching = simulated_set & target_set  # 交集
+    total_weight = 0.0
-    extra = simulated_set - target_set  # 模拟多出的产业
+    hit_weight = 0.0
-    missing = target_set - simulated_set  # 未覆盖产业
 
-    print(f"符合目标的产业数量: {len(matching)}")
+    for rank, industry in enumerate(ranked_list, start=1):
-    print(matching)
+        # 权重：排名越靠前越大（1 / rank）
-    print(f"模拟多出的产业数量: {len(extra)}")
+        w = 1.0 /rank
-    print(f"未覆盖目标产业数量: {len(missing)}")
+        total_weight += w
 
-    # ========== 6️⃣ 返回适应度（越大越好） ==========
+        if industry in target_set:
-    return (float(-error),)
+            hit_weight += w
+
+    hit_ratio = hit_weight / total_weight if total_weight > 0 else 0.0
+    error = 1.0 - hit_ratio  # GA 里：越小越好
+
+    # ---- 调试信息（强烈建议保留） ----
+    print(f"加权命中率: {hit_ratio:.4f}")
+    print(f"加权误差: {error:.4f}")
+
+    return (error,)
 # 目标产业集合
 def get_target_vulnerable_industries():
     """
@@ -156,6 +166,20 @@ def get_target_vulnerable_industries():
         '27', '8', '11', '9', '21', '96', '99', '100', '44', '45', '46', '47', '48', '49', '50', '51',
         '52', '53', '54', '55'
     ]
+    # 修改为25年的产业进行比较
+    merged_list_24 = [
+        '32', '33', '34', '35', '36', '37',
+        '38', '39', '41', '42', '43',
+        '46', '47', '48', '49',
+        '51', '52', '53', '54', '55',
+        '56', '57', '58',
+        '59', '60', '61', '62', '63', '64', '65', '66',
+        '68', '70', '71', '73', '74', '78', '79',
+        '90', '91', '92', '93', '94',
+        '95', '96', '97', '99', '100', '101',
+        '102', '103', '104', '105', '106', '107', '108', '109'
+    ]
+
     # # 提取所有 chain_id，并去重
     # chain_ids = set()
     # for item in industry_list:
@@ -166,7 +190,7 @@ def get_target_vulnerable_industries():
     #     else:
     #         chain_ids.add(str(item["chain_id"]))
 
-    return merged_list
+    return merged_list_24
 
 # 从数据库计算脆弱产业集合
 def get_vulnerable35_code(engine, ga_id):
@@ -221,7 +245,7 @@ def get_vulnerable35_code(engine, ga_id):
     # ===============================
     # 5️⃣ 选出最脆弱的前100个产品（出现次数最多）
     # ===============================
-    vulnerable100_product = count_prod.nlargest(35, "count")["id_product"].tolist()
+    vulnerable100_product = count_prod.nlargest(64, "count")["id_product"].tolist()
     print(f"[信息] ga_id={ga_id} 查询完成，共找到 {len(vulnerable100_product)} 个脆弱产品")
 
     # ===============================

GA_Agent_0925/main.py

@@ -1,178 +1,216 @@
+import csv
 import json
 import os
 import random
 from datetime import datetime
 
 from deap import tools
-from sqlalchemy.orm import close_all_sessions
 from tqdm import tqdm
 import matplotlib.pyplot as plt
 
 from GA_Agent_0925.creating import creating
-from GA_Agent_0925.orm import connection
 from controller_db import ControllerDB
-from evaluate_func import fitness, get_vulnerable35_code, get_target_vulnerable_industries
+from evaluate_func import fitness
 
-
-# ==============================
 # 遗传算法主函数（单进程）
-# ==============================
 def main():
     # 1️⃣ 加载配置
     with open("config.json", "r", encoding="utf-8") as f:
         cfg = json.load(f)
-
     random.seed(cfg["seed"])
-
     print("\n📘 参数配置:")
     for k, v in cfg.items():
         print(f"  {k}: {v}")
     print("-" * 40)
-
+    # ==============================
-    # 2️⃣ 初始化 ControllerDB（数据库连接）
+    # 2️⃣ 初始化数据库控制器
+    # ==============================
     controller_db_obj = ControllerDB("without_exp", reset_flag=0)
     controller_db_obj.reset_db(force_drop=True)
-    # 准备样本表
     controller_db_obj.prepare_list_sample()
-    # 2️⃣ 初始化工具箱
+
+    # ==============================
+    # 3️⃣ 初始化 GA
+    # ==============================
     toolbox = creating()
     pop = toolbox.population(n=cfg["pop_size"])
     hof = tools.HallOfFame(1)
+
     stats = tools.Statistics(lambda ind: ind.fitness.values)
-    stats.register("avg", lambda fits: sum(f[0] for f in fits) / len(fits))
     stats.register("max", lambda fits: max(f[0] for f in fits))
 
     best_list = []
-    avg_list = []
 
-    # ============================================================
+    # ==============================
-    # 🔧 新增内容 1：准备保存每代最优个体的文件
+    # 4️⃣ 结果保存准备
-    # ============================================================
+    # ==============================
     results_dir = "results"
     os.makedirs(results_dir, exist_ok=True)
 
-    # 文件名
     txt_result_file = os.path.join(results_dir, "best_individual_each_gen.txt")
     json_result_file = os.path.join(results_dir, "best_result_with_industry.json")
+    csv_file = "convergence0119_data.csv"
 
-    # 写入第一行：实验时间（年月日+小时）
+    exp_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
-    with open(txt_result_file, "w", encoding="utf-8") as f:
+
-        exp_time = datetime.now().strftime("%Y-%m-%d %H")
+    # TXT 初始化
-        f.write(f"实验开始时间（年月日-小时）：{exp_time}\n\n")
+    with open(txt_result_file, "a", encoding="utf-8") as f:
-        f.write("以下为每一代的最优个体基因参数：\n")
+        f.write(f"\n实验开始时间：{exp_time}\n")
+        f.write("每一代最优个体基因参数如下：\n")
+
+    # CSV 表头
+    write_header = not os.path.exists(csv_file)
+    if write_header:
+        with open(csv_file, "w", newline="", encoding="utf-8") as f:
+            writer = csv.writer(f)
+            writer.writerow([
+                "Timestamp",
+                "Generation",
+                "Best_Fitness_Percentage"
+            ])
 
     # ==============================
-    # 主进化循环
+    # 5️⃣ Fitness 缓存（关键加速）
+    # ==============================
+    fitness_cache = {}
+
+    # ==============================
+    # 6️⃣ 主进化循环
     # ==============================
     for gen in tqdm(range(cfg["n_gen"]), desc="进化中", ncols=90):
-        # 计算未评估个体适应度
+
+        # ---------- 评估当前种群 ----------
         invalid_ind = [ind for ind in pop if not ind.fitness.valid]
         for ind in invalid_ind:
+            gene_key = tuple(round(x, 4) for x in ind)
+
+            if gene_key in fitness_cache:
+                ind.fitness.values = fitness_cache[gene_key]
+            else:
                 controller_db_obj.reset_sample_db()
                 controller_db_obj.prepare_list_sample()
-            ind.fitness.values = fitness(ind, controller_db_obj=controller_db_obj)
 
-        # 选择、交叉、变异
+                fit_val = fitness(ind, controller_db_obj=controller_db_obj)
+                ind.fitness.values = fit_val
+                fitness_cache[gene_key] = fit_val
+        # ---------- 选择 ----------
         offspring = toolbox.select(pop, len(pop))
         offspring = list(map(toolbox.clone, offspring))
-
+        # ---------- 交叉 ----------
         for child1, child2 in zip(offspring[::2], offspring[1::2]):
             if random.random() < cfg["cx_prob"]:
                 toolbox.mate(child1, child2)
                 del child1.fitness.values, child2.fitness.values
-
+        # ---------- 变异 ----------
         for mutant in offspring:
             if random.random() < cfg["mut_prob"]:
                 toolbox.mutate(mutant)
                 del mutant.fitness.values
-
+        # ---------- 评估新个体 ----------
-        # 更新适应度
         invalid_ind = [ind for ind in offspring if not ind.fitness.valid]
+
         for ind in invalid_ind:
+            gene_key = tuple(round(x, 4) for x in ind)
+
+            if gene_key in fitness_cache:
+                ind.fitness.values = fitness_cache[gene_key]
+            else:
                 controller_db_obj.reset_sample_db()
                 controller_db_obj.prepare_list_sample()
-            ind.fitness.values = fitness(ind, controller_db_obj=controller_db_obj)
+
+                fit_val = fitness(ind, controller_db_obj=controller_db_obj)
+                ind.fitness.values = fit_val
+                fitness_cache[gene_key] = fit_val
 
         pop[:] = offspring
+        # ---------- 精英保留（Elitism） ----------
+        elite = hof[0] if len(hof) > 0 else None
+        if elite is not None:
+            # 替换当前种群中最差的个体（假设适应度越大越差）
+            worst_idx = max(range(len(pop)), key=lambda i: pop[i].fitness.values[0])
+            pop[worst_idx] = toolbox.clone(elite)
+
+        # 更新 hof
         hof.update(pop)
 
-        record = stats.compile(pop)
+        # ---------- 记录最优 ----------
-        best_list.append(record["max"])
-        avg_list.append(record["avg"])
-
-        # ============================================================
-        # 🔧 新增内容 2：每代实时记录最优基因到文件
-        # ============================================================
         best_ind = tools.selBest(pop, 1)[0]
+        best_fitness = best_ind.fitness.values[0]  # error，越小越好
         best_gene = list(map(float, best_ind))
-        best_ga_id = getattr(best_ind, "ga_id", None)  # 获取 ga_id，如果没有就返回 None
+        best_ga_id = getattr(best_ind, "ga_id", None)
+
+        # ---------- 维护“历史最优” ----------
+        if 'prev_best_fitness' not in globals():
+            prev_best_fitness = best_fitness
+            prev_best_gene = best_gene
+            prev_best_ga_id = best_ga_id
+        else:
+            if best_fitness < prev_best_fitness:  # 更优（误差更小）
+                prev_best_fitness = best_fitness
+                prev_best_gene = best_gene
+                prev_best_ga_id = best_ga_id
+            # 否则：保持历史最优不变
 
-        # 写入 TXT 文件
         with open(txt_result_file, "a", encoding="utf-8") as f:
             f.write(
-                (f"第 {gen + 1} 代最优基因：{best_gene}   最优适应度: {best_ind.fitness.values[0]:.4f}"
+                f"第 {gen + 1} 代 | 最优适应度: {prev_best_fitness:.4f} | 基因: {prev_best_gene}\n"
-                 if best_gene else "N/A")
-                + "\n"
             )
 
-        # ============================================================
+        # CSV
-        # 新增：删除上一轮产生的临时表
+        with open(csv_file, "a", newline="", encoding="utf-8") as f:
-        # ============================================================
+            writer = csv.writer(f)
+            writer.writerow([
+                exp_time,
+                gen,
+                best_fitness
+            ])
+
+        # ---------- 定期清理 DB ----------
+        if gen % 3 == 0:
+            controller_db_obj.drop_table(
+                "without_exp_result",
+                keep_ga_id=prev_best_ga_id
+            )
 
-        # 保留当前代最优 ga_id：
-        controller_db_obj.drop_table("without_exp_result", keep_ga_id=best_ga_id)
-        # 希望彻底删除整张表：
-        # controller_db_obj.drop_table("without_exp_result")
     # ==============================
-    # 输出最优结果
+    # 7️⃣ 输出最终结果
     # ==============================
     print("\n✅ 进化完成！")
-    print(f"🏆 最优个体: {hof[0]}")
+    print(f"🏆 最优个体: {list(map(float, hof[0]))}")
     print(f"🌟 最优适应度: {hof[0].fitness.values[0]:.4f}")
 
-    # 绘制收敛曲线
+    best_list.append(prev_best_fitness)
-    plt.figure(figsize=(12, 12))
+    # ==============================
-    plt.plot(best_list, label="Best Fitness", linewidth=2)
+    # 8️⃣ 绘制收敛曲线（仅 Best）
-    plt.plot(avg_list, label="Average Fitness", linestyle="--")
+    # ==============================
-    plt.title("Genetic Algorithm Convergence")
+    plt.figure(figsize=(12, 8))
+    plt.plot(best_list, linewidth=2, label="Best Fitness")
     plt.xlabel("Generation")
-    plt.ylabel("Fitness")
+    plt.ylabel("Error")
     plt.legend()
     plt.grid(True, alpha=0.3)
     plt.tight_layout()
-    plt.savefig("convergence1.png", dpi=300)
+    plt.savefig("convergence0119.png", dpi=300)
-    plt.show()
+    plt.close()
 
     # ==============================
-    # 最优个体产业匹配
+    # 9️⃣ 保存 JSON 结果
     # ==============================
-    print("\n📊 计算最优个体产业匹配情况...")
-
-    # ==============================
-    # 保存结果到文件
-    # ==============================
-
-    results_dir = "results"
-    os.makedirs(results_dir, exist_ok=True)
-
-    # 固定保存文件名
-    result_file = os.path.join(results_dir, "best_result_with_industry.json")
-
     result_data = {
         "config": cfg,
         "best_individual": list(map(float, hof[0])),
         "best_fitness": float(hof[0].fitness.values[0]),
         "fitness_curve": {
-            "best_list": best_list,
+            "best_list": best_list
-            "avg_list": avg_list
         },
-        "timestamp": datetime.now().strftime("%Y-%m-%d %H:%M:%S")
+        "timestamp": exp_time
     }
 
-    with open(result_file, "w", encoding="utf-8") as f:
+    with open(json_result_file, "w", encoding="utf-8") as f:
         json.dump(result_data, f, indent=4, ensure_ascii=False)
 
-    print(f"\n💾 最优结果已保存至: {result_file}")
+    print(f"\n💾 最优结果已保存至: {json_result_file}")
+
 
 if __name__ == "__main__":
     main()

GA_Agent_0925/results/best_individual_each_gen.txt

@@ -1,12 +0,0 @@
-实验开始时间（年月日-小时）：2025-11-22 14
-以下为每一代的最优个体基因参数：
-第 1 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 2 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 3 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 4 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 5 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 6 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 7 代最优基因：[51.0, 0.4550083673391433, 1.2286610916299323, 1.0, 0.06796086337575724, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 24.0, 4.322914089653634, 0.4230074859901629]   最优适应度: -31.0000
-第 8 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 9 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000
-第 10 代最优基因：[51.0, 0.4550083673391433, 0.4230074859901629, 1.0, 24.0, 0.9607789032744504, 3.0, 0.4271578472186195, 0.5342606663518962, 0.06796086337575724, 4.322914089653634, 1.2286610916299323]   最优适应度: -31.0000

GA_Agent_0925/results/best_result_with_industry.json

@@ -1,11 +1,11 @@
 {
     "config": {
-        "pop_size": 20,
+        "pop_size": 10,
-        "n_gen": 10,
+        "n_gen": 60,
         "n_var": 12,
         "bound_min": -5,
         "bound_max": 5,
-        "cx_prob": 0.5,
+        "cx_prob": 0.8,
         "mut_prob": 0.2,
         "cx_alpha": 0.5,
         "mut_sigma": 0.1,
@@ -15,45 +15,22 @@
         "seed": 42
     },
     "best_individual": [
-        51.0,
+        64.0,
-        0.4550083673391433,
+        1.764206087513454,
-        0.4230074859901629,
+        0.07889021951363398,
-        1.0,
+        0.6743314165751808,
-        24.0,
+        1.0079808674057342,
-        0.9607789032744504,
+        0.29096678837813245,
-        3.0,
+        0.3695976434649408,
-        0.4271578472186195,
+        0.2561432647676939,
-        0.5342606663518962,
+        6.0,
-        0.06796086337575724,
+        35.0,
-        4.322914089653634,
+        1.0079808674057342,
-        1.2286610916299323
+        1.0142063414994353
     ],
-    "best_fitness": -31.0,
+    "best_fitness": 0.1761567188324269,
     "fitness_curve": {
-        "best_list": [
+        "best_list": []
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0,
-            -31.0
-        ],
-        "avg_list": [
-            -32.9,
-            -32.8,
-            -32.5,
-            -32.5,
-            -32.3,
-            -32.1,
-            -32.0,
-            -32.3,
-            -32.6,
-            -32.5
-        ]
     },
-    "timestamp": "2025-11-22 19:05:08"
+    "timestamp": "2026-01-26 22:30:22"
 }

GA_Agent_0925/vulnerable35_match_results/process_vulnerable_products.py

@@ -11,9 +11,16 @@ sql_file = "query_top35_products.sql"
 
 # 指定对比列表
 specified_list = [
-    '61','68','65','66','59','71','77','7','38','95','58','90','56','57','97','98',
+    '32', '33', '34', '35', '36', '37',
-    '27','8','11','9','21','96','99','100','44','45','46','47','48','49','50','51',
+    '38', '39', '41', '42', '43',
-    '52','53','54','55'
+    '46', '47', '48', '49',
+    '51', '52', '53', '54', '55',
+    '56', '57', '58',
+    '59', '60', '61', '62', '63', '64', '65', '66',
+    '68', '70', '71', '73', '74', '78', '79',
+    '90', '91', '92', '93', '94',
+    '95', '96', '97', '99', '100', '101',
+    '102', '103', '104', '105', '106', '107', '108', '109'
 ]
 specified_set = set(specified_list)
 
@@ -35,8 +42,8 @@ print(f"[信息] 查询完成，获取到 {len(df)} 条记录。")
 all_results = []
 
 for ga_id, group in df.groupby("ga_id"):
-    vulnerable35_product = group['id_product'].astype(str).tolist()
+    vulnerable_product = group['id_product'].astype(str).tolist()
-    vulnerable_set = set(vulnerable35_product)
+    vulnerable_set = set(vulnerable_product)
 
     matching = vulnerable_set & specified_set
     extra = vulnerable_set - specified_set
@@ -44,7 +51,7 @@ for ga_id, group in df.groupby("ga_id"):
 
     all_results.append({
         "ga_id": ga_id,
-        "vulnerable35": ",".join(vulnerable35_product),
+        "vulnerable": ",".join(vulnerable_product),
         "matching_count": len(matching),
         "matching_products": ",".join(matching),
         "extra_count": len(extra),
@@ -57,5 +64,5 @@ for ga_id, group in df.groupby("ga_id"):
 # 输出 CSV
 # ===============================
 df_results = pd.DataFrame(all_results)
-df_results.to_csv("vulnerable20_match_results_new1122.csv", index=False, encoding='utf-8-sig')
+df_results.to_csv("vulnerable63_match_results_new260111.csv", index=False, encoding='utf-8-sig')
 print("[完成] 已生成 vulnerable35_match_results.csv")

GA_Agent_0925/vulnerable35_match_results/query_top35_products.sql

@@ -22,7 +22,7 @@ product_count AS (
 ),
 
 ranked_products AS (
-    -- 按每个 ga_id 的产品出现次数排序，并取前35
+    -- 按每个 ga_id 的产品出现次数排序，并取前 x
     SELECT *,
            ROW_NUMBER() OVER (PARTITION BY ga_id ORDER BY cnt DESC) AS rn
     FROM product_count

GA_Agent_0925/vulnerable35_match_results/vulnerable20_match_results_new260111.csv

@@ -0,0 +1,2 @@
+ga_id,vulnerable35,matching_count,matching_products,extra_count,extra_products,missing_count,missing_products
+1,"95,99,90,91,51,53,93,52,54,55,92,50,94,9,48,45,47,46,49,44",16,"55,49,52,95,51,53,93,90,47,54,48,99,91,94,46,92",4,"9,45,44,50",41,"109,65,66,56,34,100,74,32,42,37,35,61,64,38,68,107,73,108,57,97,59,36,103,58,60,63,96,101,71,39,62,104,105,106,70,78,79,102,33,41,43"

GA_Agent_0925/vulnerable35_match_results/vulnerable63_match_results_new260111.csv

@@ -0,0 +1,2 @@
+ga_id,vulnerable,matching_count,matching_products,extra_count,extra_products,missing_count,missing_products
+1,"95,99,90,91,51,53,93,52,54,55,92,50,94,9,48,45,47,46,49,44",16,"49,51,46,92,95,99,53,94,52,91,48,93,55,54,90,47",4,"50,44,45,9",41,"109,105,108,103,63,73,42,96,70,74,43,61,66,32,79,58,60,104,97,64,57,33,102,59,106,100,68,35,101,34,107,39,56,36,62,41,65,37,38,78,71"

GA_Agent_0925/绘图.py

@@ -0,0 +1,58 @@
+import pandas as pd
+import matplotlib.pyplot as plt
+
+# # ==============================
+# # 1️⃣ 读取 CSV
+# # ==============================
+# csv_file = "convergence0119_data.csv"
+# df = pd.read_csv(csv_file)
+#
+# # ==============================
+# # 2️⃣ 选择一次完整 GA 运行
+# # ==============================
+# target_ts = "2026-01-26 22:30:22"
+# run_df = df[df["Timestamp"] == target_ts].copy()
+#
+# # 按代数排序，取前 50 代
+# run_df = run_df.sort_values("Generation").head(100)
+
+# 读取 CSV
+csv_file = "convergence0119_data.csv"
+df = pd.read_csv(csv_file)
+
+# 按 Generation 排序
+run_df = df.sort_values("Generation")
+# ==============================
+# 3️⃣ 提取数据
+# ==============================
+generations = run_df["Generation"].values
+best_fitness = run_df["Best_Fitness_Percentage"].values
+
+# ==============================
+# 4️⃣ 计算滑动平均（窗口=5，可调）
+# ==============================
+window_size = 5
+run_df["MA"] = run_df["Best_Fitness_Percentage"].rolling(
+    window=window_size,
+    min_periods=1
+).mean()
+
+# ==============================
+# 5️⃣ 绘图：点 + 连线 + 滑动平均
+# ==============================
+plt.figure(figsize=(10, 6))
+
+# 原始点
+plt.scatter(generations, best_fitness, s=30, alpha=0.6, label="Best Error")
+
+# 连线
+plt.plot(generations, best_fitness, linewidth=1, alpha=0.6)
+
+plt.xlabel("Generation")
+plt.ylabel("Best Error")
+plt.legend()
+plt.grid(True, alpha=0.3)
+plt.tight_layout()
+
+plt.savefig("ga_convergence_me_.png", dpi=300)
+plt.show()

conf_db_prefix.yaml

@@ -1 +1 @@
-db_name_prefix: with_exp
+db_name_prefix: without_exp

conf_experiment.yaml

@@ -8,5 +8,5 @@ test:  # only for test scenarios
   n_iter: 100
 
 not_test:  # normal scenarios
-  n_sample: 10
+  n_sample: 1
   n_iter: 50

my_model.py

@@ -61,7 +61,7 @@ class MyModel(Model):
         self.cap_limit_prob_type = params['cap_limit_prob_type']  # 产能限制的概率分布类型。
         self.cap_limit_level = params['cap_limit_level']  # 产能限制的水平。
         self.diff_new_conn = params['diff_new_conn']  # 是否允许差异化的新连接。
-        # 初始化停止时间步，可能是用户通过参数传入
+        # 初始化停止时间步
         self.int_stop_ts = params.get('n_iter', 100)  # 默认停止时间为 100
 
         # 网络初始化