Refactor of scheduling for compatibility with multi-objective optimiz…

…ation (#81) - Change the contract of schedulers schedule method. It now returns a list of schedules rather than just one schedule - Pipelines now also return a list of schedules - In the Genetic Algorithm, multicriteria is now a scheduler parameter, and the standard schedule method is used to obtain schedules. This ensures compatibility of multi-objective scheduling with pipelines
aimclub · Feb 14, 2024 · 0f3c198 · 0f3c198
1 parent 40e176e
commit 0f3c198
Show file tree

Hide file tree

Showing 39 changed files with 15,026 additions and 14,727 deletions.
diff --git a/README.rst b/README.rst
@@ -130,14 +130,14 @@ To use SAMPO for the schedule generation you need to prepare:
 
       from sampo.scheduler.genetic import GeneticScheduler
 
-      scheduler = GeneticScheduler(mutate_order=0.1,
-                                   mutate_resources=0.3)
+      scheduler = GeneticScheduler(mutate_order=0.05,
+                                   mutate_resources=0.05)
 
     2.2. Schedule generation
 
     .. code-block:: python
 
-      schedule = scheduler.schedule(wg, contractors)
+      schedule = scheduler.schedule(wg, contractors)[0]
 
 3. Pipeline structure
 
@@ -151,7 +151,7 @@ When data was prepared and scheduler built, you should use scheduling pipeline t
         .wg(wg) \
         .contractors(contractors) \
         .schedule(HEFTScheduler()) \
-        .finish()
+        .finish()[0]
 
 Supported by
 ============

diff --git a/docs/source/Usage.rst b/docs/source/Usage.rst
@@ -72,14 +72,14 @@ To use SAMPO for the schedule generation you need to prepare:
 
       from sampo.scheduler.genetic import GeneticScheduler
 
-      scheduler = GeneticScheduler(mutate_order=0.1,
-                                   mutate_resources=0.3)
+      scheduler = GeneticScheduler(mutate_order=0.05,
+                                   mutate_resources=0.05)
 
     2.2. Schedule generation
 
     .. code-block:: python
 
-      schedule = scheduler.schedule(wg, contractors)
+      schedule = scheduler.schedule(wg, contractors)[0]
 
 3. Pipeline structure
 
@@ -93,4 +93,4 @@ When data was prepared and scheduler built, you should use scheduling pipeline t
         .wg(wg) \
         .contractors(contractors) \
         .schedule(HEFTScheduler()) \
-        .finish()
+        .finish()[0]
diff --git a/examples/field_development/field_development_scheduling.py b/examples/field_development/field_development_scheduling.py
@@ -46,7 +46,7 @@
 start_date = "2023-01-01"  # Set up the project's start date
 
 # Schedule field development tasks
-schedule = scheduler_type.schedule(structured_wg, contractors, validate=True)
+schedule = scheduler_type.schedule(structured_wg, contractors, validate=True)[0]
 schedule_df = schedule.merged_stages_datetime_df(start_date)
 
 # Schedule's gant chart visualization

diff --git a/examples/generator_scenarios.py b/examples/generator_scenarios.py
@@ -15,7 +15,7 @@
 p_rand = SimpleSynthetic(rand=231)
 wg = p_rand.work_graph(top_border=3000)
 contractors = [p_rand.contractor(i) for i in range(10, 31, 10)]
-schedule = HEFTScheduler().schedule(wg, contractors)
+schedule = HEFTScheduler().schedule(wg, contractors)[0]
 
 print(len(wg.nodes))
 print("\nDefault contractors")
@@ -24,7 +24,7 @@
 print("\nContractor by work graph")
 for pack_counts in []:#[1, 2, 10]:
     contractors = [get_contractor_by_wg(wg, scaler=pack_counts)]
-    execution_time = HEFTScheduler().schedule(wg, contractors).execution_time
+    execution_time = HEFTScheduler().schedule(wg, contractors)[0].execution_time
     print(f"Execution time: {execution_time}, pack count: {pack_counts}")
 
 print("\nNames extension")

diff --git a/examples/local_optimization.ipynb b/examples/local_optimization.ipynb
@@ -4,10 +4,13 @@
    "cell_type": "code",
    "execution_count": 1,
    "metadata": {
-    "collapsed": true,
     "ExecuteTime": {
      "end_time": "2023-10-31T08:06:06.215669600Z",
      "start_time": "2023-10-31T08:06:05.535679400Z"
+    },
+    "collapsed": true,
+    "jupyter": {
+     "outputs_hidden": true
     }
    },
    "outputs": [],
@@ -21,16 +24,29 @@
   },
   {
    "cell_type": "markdown",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "source": [
     "# 1. Data preparation"
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "code",
    "execution_count": 2,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-10-31T08:06:06.275585700Z",
+     "start_time": "2023-10-31T08:06:06.215669600Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "outputs": [],
    "source": [
     "# SimpleSynthetic object used for the simple work graph structure generation\n",
@@ -49,38 +65,47 @@
     "contractors = [get_contractor_by_wg(simple_wg)]\n",
     "\n",
     "scheduler = HEFTScheduler()"
-   ],
-   "metadata": {
-    "collapsed": false,
-    "ExecuteTime": {
-     "end_time": "2023-10-31T08:06:06.275585700Z",
-     "start_time": "2023-10-31T08:06:06.215669600Z"
-    }
-   }
+   ]
   },
   {
    "cell_type": "markdown",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "source": [
     "# 2. Local optimization\n",
     "There are two types of local optimization in SAMPO: order and schedule."
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "markdown",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "source": [
     "### Scheduling order optimization\n",
     "This local optimization should rearrange scheduling order to improve scheduling results."
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "code",
    "execution_count": 3,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-10-31T08:06:06.450877800Z",
+     "start_time": "2023-10-31T08:06:06.365998300Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -91,7 +116,9 @@
     },
     {
      "data": {
-      "text/plain": "1194"
+      "text/plain": [
+       "1194"
+      ]
      },
      "execution_count": 3,
      "metadata": {},
@@ -108,35 +135,43 @@
     "    .contractors(contractors) \\\n",
     "    .optimize_local(local_optimizer, range(0, 10)) \\\n",
     "    .schedule(scheduler) \\\n",
-    "    .finish()\n",
+    "    .finish()[0]\n",
     "\n",
     "project.schedule.execution_time"
-   ],
-   "metadata": {
-    "collapsed": false,
-    "ExecuteTime": {
-     "end_time": "2023-10-31T08:06:06.450877800Z",
-     "start_time": "2023-10-31T08:06:06.365998300Z"
-    }
-   }
+   ]
   },
   {
    "cell_type": "markdown",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "source": [
     "### Schedule optimization\n",
     "This local optimization should recalculate parts of schedule to make it better."
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "code",
    "execution_count": 4,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-10-31T08:06:06.635515100Z",
+     "start_time": "2023-10-31T08:06:06.465560600Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "outputs": [
     {
      "data": {
-      "text/plain": "1194"
+      "text/plain": [
+       "1194"
+      ]
      },
      "execution_count": 4,
      "metadata": {},
@@ -154,32 +189,38 @@
     "    .contractors(contractors) \\\n",
     "    .schedule(scheduler) \\\n",
     "    .optimize_local(local_optimizer, range(0, 5)) \\\n",
-    "    .finish()\n",
+    "    .finish()[0]\n",
     "\n",
     "project.schedule.execution_time"
-   ],
-   "metadata": {
-    "collapsed": false,
-    "ExecuteTime": {
-     "end_time": "2023-10-31T08:06:06.635515100Z",
-     "start_time": "2023-10-31T08:06:06.465560600Z"
-    }
-   }
+   ]
   },
   {
    "cell_type": "markdown",
+   "metadata": {
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "source": [
     "### Both\n",
     "Using pipeline you can apply both type of optimizations.\n",
     "You also can stack local optimizers, they should be applied sequentially."
-   ],
-   "metadata": {
-    "collapsed": false
-   }
+   ]
   },
   {
    "cell_type": "code",
    "execution_count": 5,
+   "metadata": {
+    "ExecuteTime": {
+     "end_time": "2023-10-31T08:06:06.860829700Z",
+     "start_time": "2023-10-31T08:06:06.690476Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
+    }
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -306,7 +347,9 @@
     },
     {
      "data": {
-      "text/plain": "1240"
+      "text/plain": [
+       "1240"
+      ]
      },
      "execution_count": 5,
      "metadata": {},
@@ -327,51 +370,47 @@
     "    .schedule(scheduler) \\\n",
     "    .optimize_local(schedule_optimizer, range(0, simple_wg.vertex_count // 2)) \\\n",
     "    .optimize_local(schedule_optimizer, range(simple_wg.vertex_count // 2, simple_wg.vertex_count)) \\\n",
-    "    .finish()\n",
+    "    .finish()[0]\n",
     "\n",
     "project.schedule.execution_time"
-   ],
-   "metadata": {
-    "collapsed": false,
-    "ExecuteTime": {
-     "end_time": "2023-10-31T08:06:06.860829700Z",
-     "start_time": "2023-10-31T08:06:06.690476Z"
-    }
-   }
+   ]
   },
   {
    "cell_type": "code",
    "execution_count": 5,
-   "outputs": [],
-   "source": [],
    "metadata": {
-    "collapsed": false,
     "ExecuteTime": {
      "end_time": "2023-10-31T08:06:06.870912400Z",
      "start_time": "2023-10-31T08:06:06.855820800Z"
+    },
+    "collapsed": false,
+    "jupyter": {
+     "outputs_hidden": false
     }
-   }
+   },
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3",
+   "display_name": "Python 3 (ipykernel)",
    "language": "python",
    "name": "python3"
   },
   "language_info": {
    "codemirror_mode": {
     "name": "ipython",
-    "version": 2
+    "version": 3
    },
    "file_extension": ".py",
    "mimetype": "text/x-python",
    "name": "python",
    "nbconvert_exporter": "python",
-   "pygments_lexer": "ipython2",
-   "version": "2.7.6"
+   "pygments_lexer": "ipython3",
+   "version": "3.12.2"
   }
  },
  "nbformat": 4,
- "nbformat_minor": 0
+ "nbformat_minor": 4
 }