`project`¶

Module Contents¶

project.datetime_stats(datetimes)¶

Gets the mean and variance of a collection of datetimes

Parameters:	datetimes (iterable(datetime)) – The datetimes to compute the statistics on.
Returns:	A dictionary containing keys for the mean and variance. The mean is a datetime, and the variance is a timedelta.
Return type:	dict

project.timedelta_stats(timedeltas)¶

Gets the mean and variance of a collection of timedeltas

Parameters:	timedeltas (iterable(timedelta)) – The timedeltas to compute the statistics on.
Returns:	A dictionary containing keys for the mean and variance. both the mean and variance are datetimes.
Return type:	dict

class project.TaskSample(duration, earliest_start, latest_finish)¶

A wrapper for a sample of the derived statistics for a Task

duration¶: timedelta – The sampled duration of the task

earliest_start¶: datetime – The sampled earliest start date of the task

latest_finish¶: datetime – The sampled latest finish date of the task

latest_start¶: datetime – The latest start date of the task. Must be set independently of the constructor

earliest_finish¶: datetime – The earliest finish date of the task. Must be set independently of the constructor.

Parameters:	duration (timedelta) – The sampled duration of the task earliest_start (datetime) – The sampled earliest start date of the task latest_finish (datetime) – The sampled latest finish date of the task

__init__(duration, earliest_start, latest_finish)¶

total_float()¶: timedelta: The total float of the task. Earliest finish mst be set before calculation.

from_task(task, current_time)¶

Constructs a TaskSample from a task

Parameters:	task (Task) – The task to sample current_time (datetime) – The current datetime used to sample the task
Returns:	The constructed sample
Return type:	TaskSample

class project.TaskStatistics(latest_start, earliest_finish, total_float)¶

A container for the relevant derived statistics for a Task

latest_start¶: dict – A dict containing the mean and variance of the latest start date of the task in ‘mean’ and ‘variance’ keys respectively.

earliest_finish¶: dict – A dict containing the mean and variance of the earliest finish date of the task in ‘mean’ and ‘variance’ keys respectively.

total_float¶: dict – A dict containing the mean and variance of the total float date of the task in ‘mean’ and ‘variance’ keys respectively.

Parameters:

latest_start (dict) – A dict containing the mean and variance of the latest start date of the task in ‘mean’ and ‘variance’ keys respectively.
earliest_finish (dict) – A dict containing the mean and variance of the earliest finish date of the task in ‘mean’ and ‘variance’ keys respectively.
total_float (dict) – A dict containing the mean and variance of the total float date of the task in ‘mean’ and ‘variance’ keys respectively.

__init__(latest_start, earliest_finish, total_float)¶

from_samples(samples)¶

Construct a TaskStatistics object from samples

Parameters:	samples (iterable(TaskSample)) – The samples to compute the statistics from.
Returns:	The constructed TaskStatistics
Return type:	TaskStatistics

__repr__()¶

class project.Project(name, model=None, uid=None, tasks=None, dependencies=None)¶

A project

Note

This must be an acyclic graph.

name¶: str – The name of the project

uid¶: UUID – The UUID of the project

model¶: A model used to predict the duration of tasks from their data

Parameters:

name (str) – The name of the project
model (optional) – A model used to predict the duration of tasks from their data
uid (UUID, optional) – The UUID of the project
tasks (iterable(Task), optional) – A collections of Tasks associated with this project
dependencies (iterable(dict), optional) – The dependencies associated with the project in the form of dicts of ‘source’ and ‘destination’ keys.

__init__(name, model=None, uid=None, tasks=None, dependencies=None)¶

validate()¶

Validates the Project meets the requirements to do inference

Checks: * The Project is a directed acyclic graph * Every terminal Task (one without successors) has a latest start date PDF

Raises:	`InvalidProject` – If the project does not conform to the requirements.

dependencies()¶: list[tuple(Task, Task)]: The dependencies in the project where the first element of the tuple is the source task and the second element of the tuple is the dependent task.

tasks()¶: iterable(Task): The tasks of this project

dependencies_summary()¶: list[DependencySummary]: The dependencies of this project

get_task_from_id(id_)¶

Gets a task from an id

Parameters:	id (UUID) – The UUID of the project to get
Returns:	The task with the associated with the id or None if task is not found
Return type:	Task\|None

add_task(task)¶

Adds a Task to this Project and determines the duration PDF of the task from the model if not previously specified.

Parameters:	task (Task) – The Task to add to the project

add_tasks(tasks)¶

Adds multiple Tasks to this Project and determines the duration PDF of the task from the model if not: previously specified.

Parameters:	tasks (iterable(Task)) – The Task to add to the project

add_dependency(parent, child)¶

Adds a Task dependency to this Project

Parameters:	parent (Task) – The parent task child (Task) – The child task, i.e. the Task which depends on the parent

add_dependencies(dependencies)¶

Adds multiple Task dependencies to this Project

Parameters:	dependencies (list[tuple(Task, Task)]) – A list of tuples of Task dependencies in the form of (parent task, child task)

calculate_earliest_finish_times(current_time=None, iterations=1000)¶

Generates samples of the earliest finish times for each uncompleted node in the project.

Parameters:	current_time (datetime) – the time at which to take the samples iterations (int, optional) – The number of samples to generate. Defaults to 1000
Returns:	[datetime]}: A dictionary of the samples for each task.
Return type:	dict{Task

earliest_finish_sample_func(parents, children, samples, **kwargs)¶

calculate_latest_start_times(iterations=1000)¶

Generates samples of the latest start times for each uncompleted node in the project.

Parameters:	iterations (int, optional) – The number of samples to generate. Defaults to 1000
Returns:	[datetime]}: A dictionary of the samples for each task.
Return type:	dict{Task

latest_start_sample_func(parents, children, samples, **kwargs)¶

_get_samples(forward_sample_func=None, backward_sample_func=None, iterations=1000, current_time=None, **kwargs)¶

_get_parents_and_children(task)¶

calculate_task_statistics(current_time=None, iterations=1000)¶

recommend_next(current_time=None, constraints=None, iterations=1000, score_func=None, selection_func=None, min_number=1, max_number=1, batch_wait=False, selection_func_arguments=None, **score_func_arguments)¶

Get the recommended next tasks

Parameters:	current_time (datetime, optional) – The current time (in UTC) to query the project. Defaults to the current time. constraints (iterable(callable)) – A list of constraints to apply to the selected tasks. These must be functions which task in two parameters – the project (self) and the set of Tasks under consideration. iterations (int, optional) – The number of iterations to query the project for each considered set of Tasks. Defaults to 1000. score_func (func, optional) – The function used to score the results of a Task set. Defaults to a function which returns a dict containing the mean and precision (inverse variance) of the total float of each task weighted by the Tasks’ deadline weight. The function must take keyword arguments which can be specified as keyword arguments to this function (see score_func_arguments). selection_func (func, optional) – The function used to select which task set is best from the results returned from the score_func. Defaults to a function which scales the total float and precision each between 0 and 1 and sums them according to a weighting parameter (see selection_func_arguments). The function must accept a dict of Task set to score and keyword arguments which can be specified by the selection_func_arguments parameter of this function. min_number (int, optional) – The minimum number of tasks which can can be recommended. Defaults to 1. max_number (int, optional) – The maximum number of tasks which can be recommended. Defaults to 1. batch_wait (bool, optional) – Do all tasks for a proposed tuple of Tasks need to be completed before the next tasks can begin? Defaults to False. selection_func_arguments (dict, optional) – The arguments to be passed to the selection_func. **score_func_arguments – The arguments to pass to the score_func
Returns:	The recommended tasks to complete next
Return type:	tuple(Task)

recommendation_sample_func(parents, children, samples, **kwargs)¶

_default_recommendation_score_func(**kwargs)¶

_default_recommendation_selection_func(**kwargs)¶

get_starting_and_terminal_tasks()¶

Gets the starting tasks (ones without predecessors) and terminal tasks (ones without successors)

Returns:	The starting and terminal tasks in the form of (starting tasks, terminal tasks)
Return type:	tuple(list[Task], list[Task])

update_from_dict(data)¶

Updates the Project using a dictionary of new values

Parameters:	data (dict) – The new values

from_dict(data_in, model)¶

Constructs a Project from a dictionary of values and a model

Parameters:	data_in (dict) – The data to construct the Project from model – The model used to predict the durations of tasks
Returns:	The constructed project
Return type:	Project

project¶

Module Contents¶

`project`¶