Visualization & import fixes

examples/field_development/field_development_scheduling.py

@@ -7,7 +7,6 @@
 from sampo.schemas.contractor import Contractor
 from sampo.schemas.graph import WorkGraph
 from sampo.structurator.base import graph_restructuring
-from sampo.utilities.schedule import remove_service_tasks
 from sampo.utilities.visualization.base import VisualizationMode
 from sampo.utilities.visualization.schedule import schedule_gant_chart_fig
 from sampo.utilities.visualization.work_graph import work_graph_fig
@@ -48,7 +47,7 @@
 
 # Schedule field development tasks
 schedule = scheduler_type.schedule(structured_wg, contractors, validate=True)
-schedule_df = remove_service_tasks(schedule.merged_stages_datetime_df(start_date))
+schedule_df = schedule.merged_stages_datetime_df(start_date)
 
 # Schedule's gant chart visualization
 gant_fig = schedule_gant_chart_fig(schedule_df,

examples/simple_synthetic_graph_scheduling.py

@@ -5,8 +5,6 @@
 
 from sampo.utilities.visualization.schedule import schedule_gant_chart_fig
 
-from sampo.utilities.schedule import remove_service_tasks
-
 from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.heft.base import HEFTScheduler
 from sampo.schemas.time import Time
@@ -47,7 +45,7 @@
 
 # Schedule works
 schedule = scheduler.schedule(wg, contractors)
-schedule_df = remove_service_tasks(schedule.merged_stages_datetime_df(start_date))
+schedule_df = schedule.merged_stages_datetime_df(start_date)
 # Schedule's gant chart visualization
 gant_fig = schedule_gant_chart_fig(schedule_df,
                                    fig_file_name=gant_chart_filename,

experiments/algo_performance_comparison.py

@@ -2,7 +2,7 @@
 
 from pathos.multiprocessing import ProcessingPool
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.generator.types import SyntheticGraphType
 from sampo.scheduler.base import SchedulerType
 from sampo.scheduler.generate import generate_schedule

experiments/algorithms_2_multi_agency.py

@@ -2,7 +2,7 @@
 from random import Random
 from typing import IO
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.heft.base import HEFTScheduler, HEFTBetweenScheduler
 from sampo.scheduler.multi_agency.block_generator import SyntheticBlockGraphType, generate_block_graph
 from sampo.scheduler.multi_agency.multi_agency import Agent, Manager

experiments/algorithms_efficiency.py

@@ -4,7 +4,7 @@
 
 from pathos.multiprocessing import ProcessingPool
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.heft.base import HEFTBetweenScheduler
 from sampo.scheduler.heft.base import HEFTScheduler
 from sampo.scheduler.multi_agency.block_generator import SyntheticBlockGraphType, generate_block_graph

experiments/genetic2baseline.py

@@ -1,6 +1,6 @@
 from uuid import uuid4
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.genetic.base import GeneticScheduler
 from sampo.schemas.contractor import Contractor
 from sampo.schemas.graph import WorkGraph

experiments/genetic_2_multi_agency.py

@@ -1,6 +1,6 @@
 from random import Random
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.genetic.base import GeneticScheduler
 from sampo.scheduler.multi_agency.block_generator import SyntheticBlockGraphType, generate_block_graph
 from sampo.scheduler.multi_agency.multi_agency import Agent, Manager

experiments/modular_examples/multi_agency_comparison.py

@@ -2,7 +2,7 @@
 from random import Random
 from typing import IO
 
-from sampo.generator import SimpleSynthetic
+from sampo.generator.base import SimpleSynthetic
 from sampo.scheduler.base import Scheduler
 from sampo.scheduler.genetic.base import GeneticScheduler
 from sampo.scheduler.heft.base import HEFTBetweenScheduler

pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "sampo"
-version = "0.1.1.225"
+version = "0.1.1.231"
 description = "Open-source framework for adaptive manufacturing processes scheduling"
 authors = ["iAirLab <[email protected]>"]
 license = "BSD-3-Clause"
@@ -11,16 +11,15 @@ license = "BSD-3-Clause"
 [tool.poetry.dependencies]
 python = ">=3.10,<3.11"
 sortedcontainers = ">=2.4.0,<2.5.0"
-numpy = ">=1.23.5,<1.24.0"
-pandas = ">=1.5.2,<1.6.0"
+numpy = ">=1.23.5"
+pandas = ">=2.0.0"
 scipy = ">=1.9.3,<1.10.0"
 toposort = ">=1.7,<2.0"
 deap = ">=1.3.3,<1.4.0"
 seaborn = ">=0.12.1,<0.13.0"
 matplotlib = ">=3.6.2,<3.7.0"
-plotly = ">=5.11.0,<5.12.0"
+plotly = ">=5.17.0,<5.18.0"
 pytest = ">=7.2.0,<7.3.0"
-pathos = ">=0.3.0,<0.3.1"
 
 
 [build-system]

requirements.txt

@@ -1,12 +1,12 @@
 sortedcontainers~=2.4.0
 numpy~=1.23.5
-pandas~=1.5.2
+pandas~=2.0.0
 scipy~=1.9.3
 toposort~=1.7
 deap~=1.3.3
 seaborn~=0.12.1
 matplotlib~=3.6.2
-plotly~=5.11.0
+plotly~=5.17.0
 pytest~=7.2.0
 pytest-xdist~=3.1.0
 pathos~=0.3.0

sampo/schemas/schedule.py

@@ -106,7 +106,7 @@ def merged_stages_datetime_df(self, offset: Union[datetime, str]) -> DataFrame:
         :param offset: Start of schedule, to add as an offset.
         :return: Shifted schedule DataFrame with merged tasks.
         """
-        result = fix_split_tasks(self.offset_schedule(offset))
+        result = self.offset_schedule(offset)
         return result
 
     def offset_schedule(self, offset: Union[datetime, str]) -> DataFrame:
@@ -160,7 +160,7 @@ def sed(time1, time2) -> tuple:
                        ) for i, w in enumerate(works)]
         data_frame = DataFrame.from_records(data_frame, columns=Schedule._columns)
 
-        data_frame = data_frame.set_index('idx')
+        data_frame = data_frame.set_index('idx', drop=False)
 
         if ordered_task_ids:
             data_frame.task_id = data_frame.task_id.astype('category')

sampo/schemas/time_estimator.py

@@ -6,6 +6,7 @@
 
 import numpy.random
 
+from sampo.schemas.requirements import WorkerReq
 from sampo.schemas.resources import Worker
 from sampo.schemas.resources import WorkerProductivityMode
 from sampo.schemas.time import Time
@@ -33,7 +34,8 @@ def set_productivity_mode(self, mode: WorkerProductivityMode = WorkerProductivit
         ...
 
     @abstractmethod
-    def find_work_resources(self, work_name: str, work_volume: float, resource_name: list[str] = None) -> dict[str, int]:
+    def find_work_resources(self, work_name: str, work_volume: float, resource_name: list[str] | None = None) \
+            -> list[WorkerReq]:
         ...
 
     @abstractmethod
@@ -51,10 +53,14 @@ def __init__(self,
         self._productivity_mode = WorkerProductivityMode.Static
 
     def find_work_resources(self, work_name: str, work_volume: float, resource_name: list[str] | None = None) \
-            -> dict[str, int]:
+            -> list[WorkerReq]:
         if resource_name is None:
             resource_name = ['driver', 'fitter', 'manager', 'handyman', 'electrician', 'engineer']
-        return dict((name, numpy.random.poisson(work_volume ** 0.5, 1)[0]) for name in resource_name)
+        return [WorkerReq(kind=name,
+                          volume=work_volume * numpy.random.poisson(work_volume ** 0.5, 1)[0],
+                          min_count=numpy.random.poisson(work_volume ** 0.2, 1)[0],
+                          max_count=numpy.random.poisson(work_volume * 3, 1)[0])
+                for name in resource_name]
 
     def set_estimation_mode(self, use_idle: bool = True, mode: WorkEstimationMode = WorkEstimationMode.Realistic):
         self._use_idle = use_idle

sampo/userinput/parser/csv_parser.py

@@ -125,7 +125,8 @@ def work_graph_and_contractors(works_info: pd.DataFrame,
             works_info.activity_name = works_info.activity_name.apply(lambda name: unique_work_names_mapper[name])
 
         if contractor_info is None:
-            resources = [work_resource_estimator.find_work_resources(w[0], float(w[1]))
+            resources = [dict((worker_req.name, int(worker_req.volume))
+                              for worker_req in work_resource_estimator.find_work_resources(w[0], float(w[1])))
                          for w in works_info.loc[:, ['activity_name', 'volume']].to_numpy()]
             contractors = [get_contractor_for_resources_schedule(resources,
                                                                  contractor_capacity=contractor_types[i],
@@ -134,7 +135,8 @@ def work_graph_and_contractors(works_info: pd.DataFrame,
                            for i in range(contractors_number)]
         elif isinstance(contractor_info, list):
             contractors = contractor_info
-            resources = [work_resource_estimator.find_work_resources(w[0], float(w[1]))
+            resources = [dict((worker_req.name, int(worker_req.volume))
+                              for worker_req in work_resource_estimator.find_work_resources(w[0], float(w[1])))
                          for w in works_info.loc[:, ['activity_name', 'volume']].to_numpy()]
         else:
             # if contractor info is given or contractor info and work resource estimator are received simultaneously
@@ -152,11 +154,8 @@ def work_graph_and_contractors(works_info: pd.DataFrame,
                                equipments=dict())
                 )
             resource_names = contractor_df.columns[1:].to_list()
-            if len(contractors) == 0 and isinstance(work_resource_estimator, DefaultWorkEstimator):
-                raise InputDataException(
-                    'you have neither info about contractors nor work resource estimator.'
-                )
-            resources = [work_resource_estimator.find_work_resources(w[0], float(w[1]), resource_names)
+            resources = [dict((worker_req.name, int(worker_req.volume))
+                              for worker_req in work_resource_estimator.find_work_resources(w[0], float(w[1]), resource_names))
                          for w in works_info.loc[:, ['activity_name', 'volume']].to_numpy()]
 
         unique_res = list(set(chain(*[r.keys() for r in resources])))

sampo/utilities/schedule.py

@@ -20,7 +20,7 @@ def fix_split_tasks(baps_schedule_df: pd.DataFrame) -> pd.DataFrame:
         task_stages_df = baps_schedule_df.loc[
             baps_schedule_df.task_id.str.startswith(f'{task_id}{STAGE_SEP}')
             | (baps_schedule_df.task_id == task_id)
-        ]
+            ]
         task_series = merge_split_stages(task_stages_df.reset_index(drop=True))
         df.loc[df.shape[0]] = task_series  # append
 
@@ -37,14 +37,11 @@ def merge_split_stages(task_df: pd.DataFrame) -> pd.Series:
     :param task_df: pd.DataFrame: one real task's stages dataframe, sorted by start time
     :return: pd.Series with the full information about the task
     """
-    if len(task_df) == 1:
-        df = task_df.copy()
-        df['successors'] = [[tuple([x[0].split(STAGE_SEP)[0], x[1]]) for x in df.loc[0, 'successors']]]
-        return df.loc[0, :]
-    else:
+    if len(task_df) > 1:
         task_df = task_df.sort_values(by='task_name_mapped')
         task_df = task_df.reset_index(drop=True)
         df = task_df.copy()
+
         df = df.iloc[-1:].reset_index(drop=True)
         for column in ['task_id', 'task_name']:
             df.loc[0, column] = df.loc[0, column].split(STAGE_SEP)[0]  # fix task id and name
@@ -53,46 +50,11 @@ def merge_split_stages(task_df: pd.DataFrame) -> pd.Series:
         df.loc[0, 'volume'] = sum(task_df.loc[:, 'volume'])
         df.loc[0, 'workers'] = task_df.loc[0, 'workers']
 
-        # fix connections through all stages
-        fixed_connections_lst = []
-        for connections_lst in task_df.loc[:, 'successors']:
-            for connection in connections_lst:
-                if connection[1] != 'IFS':
-                    fixed_connections_lst.append(tuple([connection[0].split('_')[0], connection[1]]))
-        fixed_connections_lst = list(set(fixed_connections_lst))
-        df.loc[:, 'successors'] = [fixed_connections_lst]
-
         # fix task's start time and duration
         df.loc[0, 'start'] = task_df.loc[0, 'start']
         df.loc[0, 'finish'] = task_df.loc[len(task_df) - 1, 'finish']
         df.loc[0, 'duration'] = (df.loc[0, 'finish'] - df.loc[0, 'start']).days + 1
+    else:
+        df = task_df.copy()
 
-        return df.loc[0, :]
-
-
-def remove_service_tasks(service_schedule_df: pd.DataFrame) -> pd.DataFrame:
-    """
-    Remove 'start', 'finish' and milestone tasks from the schedule
-
-    :param service_schedule_df: pd.DataFrame: schedule (with merges stages in the case of baps) with service tasks
-    :return: pd.DataFrame: schedule without information about service tasks
-    """
-    schedule_df = service_schedule_df.copy()
-
-    service_df = schedule_df.loc[:, 'task_name'].str.contains('start|finish')
-
-    # Prepare list with service tasks ids
-    service_tasks_ids = set(schedule_df.loc[service_df].loc[:, 'task_id'])
-
-    # Remove rows with service tasks from DataFrame
-    schedule_df = schedule_df.loc[~service_df]
-
-    # Fix connections linked to the service tasks
-    fixed_connections_lst = []
-    for connections_lst in schedule_df.loc[:, 'successors']:
-        fixed_connections_lst.append([])
-        for connection in connections_lst:
-            if connection[0] not in service_tasks_ids:
-                fixed_connections_lst[-1].append(connection)
-    schedule_df.loc[:, 'successors'] = pd.Series(fixed_connections_lst)
-    return schedule_df
+    return df.loc[0, :]

sampo/utilities/visualization/schedule.py

@@ -32,37 +32,41 @@ def schedule_gant_chart_fig(schedule_dataframe: pd.DataFrame,
     visualization_start_delta = timedelta(days=2)
     visualization_finish_delta = timedelta(days=(schedule_finish - schedule_start).days // 3)
 
-    def create_zone_row(i, work_name, zone_names, zone) -> dict:
+    def create_zone_row(i, zone_names, zone) -> dict:
         return {'idx': i,
                 'contractor': 'Access cards',
                 'cost': 0,
                 'volume': 0,
                 'duration': 0,
                 'measurement': 'unit',
-                'successors': [],
                 'workers_dict': '',
                 'workers': '',
                 'task_name_mapped': zone_names,
                 'task_name': '',
-                'start': timedelta(int(zone.start_time)) + schedule_start - visualization_start_delta + timedelta(1),
-                'finish': timedelta(int(zone.end_time)) + schedule_start  - visualization_start_delta + timedelta(1)}
+                'zone_information': '',
+                'start': timedelta(int(zone.start_time)) + schedule_start,
+                'finish': timedelta(int(zone.end_time)) + schedule_start}
 
     sworks = schedule_dataframe['scheduled_work_object'].copy()
     idx = schedule_dataframe['idx'].copy()
 
     def get_zone_usage_info(swork) -> str:
-        return '<br>' + '<br>'.join([f'{zone.kind}: {zone.required_status}' for zone in swork.work_unit.zone_reqs])
+        return '<br>' + '<br>'.join([f'{zone.name}: {zone.to_status}' for zone in swork.zones_pre])
 
     schedule_dataframe['zone_information'] = sworks.apply(get_zone_usage_info)
 
+    access_cards = []
+
     # create zone information
     for i, swork in zip(idx, sworks):
         zone_names = '<br>' + '<br>'.join([zone.name for zone in swork.zones_pre])
         for zone in swork.zones_pre:
-            schedule_dataframe = schedule_dataframe.append(create_zone_row(i, swork.work_unit.name, zone_names, zone), ignore_index=True)
+            access_cards.append(create_zone_row(i, zone_names, zone))
         zone_names = '<br>' + '<br>'.join([zone.name for zone in swork.zones_post])
         for zone in swork.zones_post:
-            schedule_dataframe = schedule_dataframe.append(create_zone_row(i, swork.work_unit.name, zone_names, zone), ignore_index=True)
+            access_cards.append(create_zone_row(i, zone_names, zone))
+
+    schedule_dataframe = pd.concat([schedule_dataframe, pd.DataFrame.from_records(access_cards)])
 
     schedule_dataframe['color'] = schedule_dataframe[['task_name', 'contractor']] \
         .apply(lambda r: 'Defect' if ':' in r['task_name'] else r['contractor'], axis=1)