Adds documentation for some of the classes and methods #31

zaouk · 2022-04-05T09:20:17Z

No description provided.

juanmc2005

Thanks for this very needed documentation!
In general I prefer docstrings to follow the same format as in AggregationStrategy.__call__ (see here) so before I merge I'd like to reformat these docstrings.

Apart from that I added some suggestions that in my opinion are a bit clearer.

juanmc2005 · 2022-04-07T13:53:00Z

src/diart/functional.py

@@ -288,6 +288,19 @@ def __init__(
        metric: Optional[str] = "cosine",
        max_speakers: int = 20
    ):
+        """Initializes an object for constrained incremental online clustering of speakers


I'd like to put this docstring at the level of the class instead of the constructor.

Also I prefer this definition:

Suggested change

"""Initializes an object for constrained incremental online clustering of speakers

"""Constrained online speaker clustering for global speaker tracking.

juanmc2005 · 2022-04-07T13:55:11Z

src/diart/functional.py

+        """Initializes an object for constrained incremental online clustering of speakers
+
+        Args:
+            tau_active (float): Threshold for detecting active speakers. This threshold is applied on the maximum value of each output


Suggested change

tau_active (float): Threshold for detecting active speakers. This threshold is applied on the maximum value of each output

tau_active (float): Threshold for detecting active speakers. This is applied on the maximum value of per-speaker segmentation

juanmc2005 · 2022-04-07T13:55:24Z

src/diart/functional.py

+
+        Args:
+            tau_active (float): Threshold for detecting active speakers. This threshold is applied on the maximum value of each output
+                                activation of the local segmentation model.


Suggested change

activation of the local segmentation model.

activations from the local segmentation model.

juanmc2005 · 2022-04-07T13:57:08Z

src/diart/functional.py

+            tau_active (float): Threshold for detecting active speakers. This threshold is applied on the maximum value of each output
+                                activation of the local segmentation model.
+            rho_update (float): Threshold for considering the extracted embedding when updating the centroid of the local speaker.
+                                The centroid to which a local speaker is mapped is only updated if the duration of speech of a


Suggested change

The centroid to which a local speaker is mapped is only updated if the duration of speech of a

The centroid to which a local speaker is mapped is only updated if its ratio of speech

juanmc2005 · 2022-04-07T13:57:22Z

src/diart/functional.py

+                                activation of the local segmentation model.
+            rho_update (float): Threshold for considering the extracted embedding when updating the centroid of the local speaker.
+                                The centroid to which a local speaker is mapped is only updated if the duration of speech of a
+                                local speaker is greater than this threshold.


Suggested change

local speaker is greater than this threshold.

is greater than this threshold.

juanmc2005 · 2022-04-07T14:13:31Z

src/diart/mapping.py

+            assignments (Iterable[Tuple[int, int]]): An iterable of tuples for assigning each local speaker to one global speaker.
+
+        Returns:
+            SpeakerMap: A SpeakerMap object wrapping a mapping matrix between local speakers and global speakers.


Suggested change

SpeakerMap: A SpeakerMap object wrapping a mapping matrix between local speakers and global speakers.

a `SpeakerMap` with the given assignments

juanmc2005 · 2022-04-07T14:15:04Z

src/diart/mapping.py

@@ -82,6 +92,16 @@ class SpeakerMapBuilder:
    def hard_map(
        shape: Tuple[int, int], assignments: Iterable[Tuple[int, int]], maximize: bool
    ) -> SpeakerMap:
+        """Returns a SpeakerMap object based on the given assignments.


Suggested change

"""Returns a SpeakerMap object based on the given assignments.

"""Builds a `SpeakerMap` that forces the given assignments.

juanmc2005 · 2022-04-07T14:15:48Z

src/diart/mapping.py

+
+        Args:
+            shape (Tuple[int, int]): shape of the mapping matrix
+            assignments (Iterable[Tuple[int, int]]): An iterable of tuples for assigning each local speaker to one global speaker.


Suggested change

assignments (Iterable[Tuple[int, int]]): An iterable of tuples for assigning each local speaker to one global speaker.

assignments (Iterable[Tuple[int, int]]): Iterable of tuples (source_speaker, target_speaker) representing an assignment.

juanmc2005 · 2022-04-07T14:16:54Z

src/diart/mapping.py

+        Args:
+            shape (Tuple[int, int]): shape of the mapping matrix
+            assignments (Iterable[Tuple[int, int]]): An iterable of tuples for assigning each local speaker to one global speaker.
+            maximize (bool): whether or not to use a MaximizationObjective


Suggested change

maximize (bool): whether or not to use a MaximizationObjective

maximize (bool): Whether to use scores where higher is better (true) or where lower is better (false).

juanmc2005 · 2022-04-07T14:17:28Z

src/diart/mapping.py

+            maximize (bool): whether or not to use a MaximizationObjective
+
+        Returns:
+            SpeakerMap: A SpeakerMap object.


Suggested change

SpeakerMap: A SpeakerMap object.

a `SpeakerMap` with the given assignments.

…al.py and mapping.py

juanmc2005

Awesome, just some style changes that I'll commit before merging.

src/diart/functional.py

src/diart/mapping.py

juanmc2005 requested review from juanmc2005 April 7, 2022 11:30

juanmc2005 added documentation Improvements or additions to documentation labels Apr 7, 2022

juanmc2005 modified the milestone: Version 0.3 Apr 7, 2022

juanmc2005 changed the base branch from main to develop April 7, 2022 13:44

juanmc2005 requested changes Apr 7, 2022

View reviewed changes

juanmc2005 assigned zaouk May 6, 2022

juanmc2005 removed this from the Version 0.3 milestone May 12, 2022

juanmc2005 added this to the Version 0.4 milestone May 20, 2022

Adds documentation for some of the classes and methods under function…

18983aa

…al.py and mapping.py

zaouk force-pushed the documentation branch from 7cf7234 to 18983aa Compare June 20, 2022 08:47

juanmc2005 approved these changes Jun 20, 2022

View reviewed changes

juanmc2005 added 12 commits June 20, 2022 11:02

Update src/diart/functional.py

c420da0

Update src/diart/functional.py

259c1c2

Update src/diart/functional.py

2a79cef

Update src/diart/functional.py

80fb42f

Update src/diart/functional.py

d73a612

Update src/diart/functional.py

b7c5ce5

Update src/diart/mapping.py

a235b82

Update src/diart/mapping.py

d715ac8

Update src/diart/mapping.py

a3e253e

Update src/diart/functional.py

85a5e16

Update src/diart/functional.py

2c759cc

Update src/diart/functional.py

b447148

juanmc2005 added 6 commits June 20, 2022 11:05

Update src/diart/functional.py

9578743

Update src/diart/functional.py

eb59d04

Update src/diart/mapping.py

1b0f5c1

Update src/diart/mapping.py

7f2b9d0

Update src/diart/mapping.py

547edb3

Include latest changes from develop

1966b3d

juanmc2005 merged commit a6ddb07 into juanmc2005:develop Jun 20, 2022

juanmc2005 mentioned this pull request Jun 20, 2022

Version 0.4 #66

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adds documentation for some of the classes and methods #31

Adds documentation for some of the classes and methods #31

zaouk commented Apr 5, 2022

juanmc2005 left a comment

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 Apr 7, 2022

juanmc2005 left a comment

	"""Initializes an object for constrained incremental online clustering of speakers
	"""Constrained online speaker clustering for global speaker tracking.

	tau_active (float): Threshold for detecting active speakers. This threshold is applied on the maximum value of each output
	tau_active (float): Threshold for detecting active speakers. This is applied on the maximum value of per-speaker segmentation

	activation of the local segmentation model.
	activations from the local segmentation model.

	The centroid to which a local speaker is mapped is only updated if the duration of speech of a
	The centroid to which a local speaker is mapped is only updated if its ratio of speech

	local speaker is greater than this threshold.
	is greater than this threshold.

	SpeakerMap: A SpeakerMap object wrapping a mapping matrix between local speakers and global speakers.
	a `SpeakerMap` with the given assignments

	"""Returns a SpeakerMap object based on the given assignments.
	"""Builds a `SpeakerMap` that forces the given assignments.

	assignments (Iterable[Tuple[int, int]]): An iterable of tuples for assigning each local speaker to one global speaker.
	assignments (Iterable[Tuple[int, int]]): Iterable of tuples (source_speaker, target_speaker) representing an assignment.

	maximize (bool): whether or not to use a MaximizationObjective
	maximize (bool): Whether to use scores where higher is better (true) or where lower is better (false).

	SpeakerMap: A SpeakerMap object.
	a `SpeakerMap` with the given assignments.

Adds documentation for some of the classes and methods #31

Adds documentation for some of the classes and methods #31

Conversation

zaouk commented Apr 5, 2022

juanmc2005 left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

juanmc2005 left a comment

Choose a reason for hiding this comment