job-shop-lib 0.1.3__tar.gz → 0.2.0__tar.gz

{job_shop_lib-0.1.3 → job_shop_lib-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: job-shop-lib
-Version: 0.1.3
+Version: 0.2.0
 Summary: An easy-to-use and modular Python library for the Job Shop Scheduling Problem (JSSP)
 License: MIT
 Author: Pabloo22
@@ -33,8 +33,6 @@ Description-Content-Type: text/markdown
 
 </div>
 
-
-
 An easy-to-use and modular Python library for the Job Shop Scheduling Problem (JSSP) with a special focus on graph representations.
 
 It provides intuitive data structures to represent instances and solutions, as well as solvers and visualization tools.
@@ -174,7 +172,7 @@ We can visualize the solution with a `DispatchingRuleSolver` as a gif:
 
 ```python
 from job_shop_lib.visualization import create_gif, get_plot_function
-from job_shop_lib.solvers import DispatchingRuleSolver, DispatchingRule
+from job_shop_lib.dispatching import DispatchingRuleSolver, DispatchingRule
 
 plt.style.use("ggplot")
 

{job_shop_lib-0.1.3 → job_shop_lib-0.2.0}/README.md

@@ -11,8 +11,6 @@
 
 </div>
 
-
-
 An easy-to-use and modular Python library for the Job Shop Scheduling Problem (JSSP) with a special focus on graph representations.
 
 It provides intuitive data structures to represent instances and solutions, as well as solvers and visualization tools.
@@ -152,7 +150,7 @@ We can visualize the solution with a `DispatchingRuleSolver` as a gif:
 
 ```python
 from job_shop_lib.visualization import create_gif, get_plot_function
-from job_shop_lib.solvers import DispatchingRuleSolver, DispatchingRule
+from job_shop_lib.dispatching import DispatchingRuleSolver, DispatchingRule
 
 plt.style.use("ggplot")
 

job_shop_lib-0.2.0/job_shop_lib/generators/transformations.py

@@ -0,0 +1,164 @@
+"""Classes for generating transformed JobShopInstance objects."""
+
+import abc
+import copy
+import random
+
+from job_shop_lib import JobShopInstance, Operation
+
+
+class Transformation(abc.ABC):
+    """Base class for transformations applied to JobShopInstance objects."""
+
+    def __init__(self, suffix: str = ""):
+        self.suffix = suffix
+        self.counter = 0
+
+    @abc.abstractmethod
+    def apply(self, instance: JobShopInstance) -> JobShopInstance:
+        """Applies the transformation to a given JobShopInstance.
+
+        Args:
+            instance: The JobShopInstance to transform.
+
+        Returns:
+            A new JobShopInstance with the transformation applied.
+        """
+
+    def __call__(self, instance: JobShopInstance) -> JobShopInstance:
+        instance = self.apply(instance)
+        suffix = f"{self.suffix}_id={self.counter}"
+        instance.name += suffix
+        self.counter += 1
+        return instance
+
+
+# pylint: disable=too-few-public-methods
+class RemoveMachines(Transformation):
+    """Removes operations associated with randomly selected machines until
+    there are exactly num_machines machines left."""
+
+    def __init__(self, num_machines: int, suffix: str | None = None):
+        if suffix is None:
+            suffix = f"_machines={num_machines}"
+        super().__init__(suffix=suffix)
+        self.num_machines = num_machines
+
+    def apply(self, instance: JobShopInstance) -> JobShopInstance:
+        if instance.num_machines <= self.num_machines:
+            return instance  # No need to remove machines
+
+        # Select machine indices to keep
+        machines_to_keep = set(
+            random.sample(range(instance.num_machines), self.num_machines)
+        )
+
+        # Re-index machines
+        machine_reindex_map = {
+            old_id: new_id
+            for new_id, old_id in enumerate(sorted(machines_to_keep))
+        }
+
+        new_jobs = []
+        for job in instance.jobs:
+            # Keep operations whose machine_id is in machines_to_keep and
+            # re-index them
+            new_jobs.append(
+                [
+                    Operation(machine_reindex_map[op.machine_id], op.duration)
+                    for op in job
+                    if op.machine_id in machines_to_keep
+                ]
+            )
+
+        return JobShopInstance(new_jobs, instance.name)
+
+
+# pylint: disable=too-few-public-methods
+class AddDurationNoise(Transformation):
+    """Adds uniform integer noise to operation durations."""
+
+    def __init__(
+        self,
+        min_duration: int = 1,
+        max_duration: int = 100,
+        noise_level: int = 10,
+        suffix: str | None = None,
+    ):
+        if suffix is None:
+            suffix = f"_noise={noise_level}"
+        super().__init__(suffix=suffix)
+        self.min_duration = min_duration
+        self.max_duration = max_duration
+        self.noise_level = noise_level
+
+    def apply(self, instance: JobShopInstance) -> JobShopInstance:
+        new_jobs = []
+        for job in instance.jobs:
+            new_job = []
+            for op in job:
+                noise = random.randint(-self.noise_level, self.noise_level)
+                new_duration = max(
+                    self.min_duration,
+                    min(self.max_duration, op.duration + noise),
+                )
+
+                new_job.append(Operation(op.machine_id, new_duration))
+            new_jobs.append(new_job)
+
+        return JobShopInstance(new_jobs, instance.name)
+
+
+class RemoveJobs(Transformation):
+    """Removes jobs randomly until the number of jobs is within a specified
+    range."""
+
+    def __init__(
+        self,
+        min_jobs: int,
+        max_jobs: int,
+        target_jobs: int | None = None,
+        suffix: str | None = None,
+    ):
+        """
+        Args:
+            min_jobs: The minimum number of jobs to remain in the instance.
+            max_jobs: The maximum number of jobs to remain in the instance.
+            target_jobs: If specified, the number of jobs to remain in the
+                instance. Overrides min_jobs and max_jobs.
+        """
+        if suffix is None:
+            suffix = f"_jobs={min_jobs}-{max_jobs}"
+        super().__init__(suffix=suffix)
+        self.min_jobs = min_jobs
+        self.max_jobs = max_jobs
+        self.target_jobs = target_jobs
+
+    def apply(self, instance: JobShopInstance) -> JobShopInstance:
+        if self.target_jobs is None:
+            target_jobs = random.randint(self.min_jobs, self.max_jobs)
+        else:
+            target_jobs = self.target_jobs
+        new_jobs = copy.deepcopy(instance.jobs)
+
+        while len(new_jobs) > target_jobs:
+            new_jobs.pop(random.randint(0, len(new_jobs) - 1))
+
+        return JobShopInstance(new_jobs, instance.name)
+
+    @staticmethod
+    def remove_job(
+        instance: JobShopInstance, job_index: int
+    ) -> JobShopInstance:
+        """Removes a specific job from the instance.
+
+        Args:
+            instance: The JobShopInstance from which to remove the job.
+            job_index: The index of the job to remove.
+
+        Returns:
+            A new JobShopInstance with the specified job removed.
+        """
+        new_jobs = copy.deepcopy(instance.jobs)
+        new_jobs.pop(job_index)
+        return JobShopInstance(new_jobs, instance.name)

{job_shop_lib-0.1.3 → job_shop_lib-0.2.0}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "job-shop-lib"
-version = "0.1.3"
+version = "0.2.0"
 description = "An easy-to-use and modular Python library for the Job Shop Scheduling Problem (JSSP)"
 authors = ["Pabloo22 <pablete.arino@gmail.com>"]
 license = "MIT"