Release notes v1.0.0

docs/scripts/make_model_zoo_docs.py

@@ -90,13 +90,9 @@
     import fiftyone.zoo as foz
 {% if 'segment-anything' in name and 'video' in name %}
     from fiftyone import ViewField as F
-{% endif %}
-
-{% if 'med-sam' in name %}
+{% elif 'med-sam' in name %}
     from fiftyone import ViewField as F
     from fiftyone.utils.huggingface import load_from_hub
-
-    dataset = load_from_hub("Voxel51/BTCV-CT-as-video-MedSAM2-dataset")[:2]
 {% endif %}
 
 {% if 'imagenet' in name %}
@@ -117,6 +113,7 @@
         .save()
     )
 {% elif 'med-sam' in name %}
+    dataset = load_from_hub("Voxel51/BTCV-CT-as-video-MedSAM2-dataset")[:2]
 
     # Retaining detections from a single frame in the middle
     # Note that SAM2 only propagates segmentation masks forward in a video
@@ -126,7 +123,6 @@
         .set_field("frames.gt_detections", None)
         .save()
     )
-
 {% else %}
     dataset = foz.load_zoo_dataset(
         "coco-2017",

docs/source/deprecation.rst

@@ -5,19 +5,23 @@ FiftyOne Deprecation Notices
 
 .. default-role:: code
 
+.. _deprecation-fiftyone-desktop:
+
 FiftyOne Desktop
 ----------------
-*Support ended with 0.25.0*
+*Support ended with FiftyOne 0.25.0*
 
-A compatible `fiftyone-desktop https://pypi.org/project/fiftyone-desktop/`_
+A compatible `fiftyone-desktop <https://pypi.org/project/fiftyone-desktop>`_
 package is no longer available as of `fiftyone==0.25.0`.
 
 Chromium-based browsers, Firefox, or a :ref:`notebook <notebooks>` environment
 are recommended for the best FiftyOne experience.
 
+.. _deprecation-python-3.8:
+
 Python 3.8
 ----------
-*Support Ended October 2024*
+*Support ended October 1, 2024*
 
 `Python 3.8 <https://devguide.python.org/versions/>`_
 transitions to `end-of-life` effective October of 2024. FiftyOne releases after

docs/source/index.rst

@@ -181,6 +181,11 @@ learn how:
     :image_src: https://voxel51.com/images/integrations/v7-128.png
     :image_title: V7
 
+.. customimagelink::
+    :image_link: https://github.com/segments-ai/segments-voxel51-plugin
+    :image_src: https://voxel51.com/images/integrations/segments-128.png
+    :image_title: Segments
+
 .. customimagelink::
     :image_link: integrations/labelbox.html
     :image_src: https://voxel51.com/images/integrations/labelbox-128.png

docs/source/plugins/developing_plugins.rst

@@ -1774,12 +1774,11 @@ in the App.
 Panels can be defined in either Python or JS, and FiftyOne comes with a
 number of :ref:`builtin panels <plugins-design-panels>` for common tasks.
 
-Depending on the ``surfaces`` panel config, panels can be scoped to either
-the grid or the modal. You can open these panels from the "+" menu, which
-is available in both the grid and modal views. Whereas grid panels enable
-extensibility at the macro level, allowing you to work with entire datasets,
-modal panels provide extensibility at the micro level, focusing on individual
-samples and scenarios.
+Panels can be scoped to the App's grid view or modal view via their
+:ref:`config <panel-config>`. Grid panels enable extensibility at the macro
+level, allowing you to work with entire datasets or views, while modal panels
+provide extensibility at the micro level, focusing on individual samples and
+scenarios.
 
 Panels, like :ref:`operators <developing-operators>`, can make use of the
 :mod:`fiftyone.operators.types` module and the
@@ -1838,10 +1837,9 @@ subsequent sections.
                 # Whether to allow multiple instances of the panel to be opened
                 allow_multiple=False,
 
-                # Whether the panel should be available in the grid view
-                # modal view, or both
+                # Whether the panel should be available in the grid, modal, or both
                 # Possible values: "grid", "modal", "grid modal"       
-                surfaces="grid modal" # default = "grid"
+                surfaces="grid",  # default = "grid"
 
                 # Markdown-formatted text that describes the panel. This is
                 # rendererd in a tooltip when the help icon in the panel
@@ -2138,7 +2136,7 @@ Panel config
 
 Every panel must define a
 :meth:`config <fiftyone.operators.panel.Panel.config>` property that
-defines its name, display name, and other optional metadata about its
+defines its name, display name, surfaces, and other optional metadata about its
 behavior:
 
 .. code-block:: python
@@ -2162,8 +2160,26 @@ behavior:
 
             # Whether to allow multiple instances of the panel to be opened
             allow_multiple=False,
+
+            # Whether the panel should be available in the grid, modal, or both
+            # Possible values: "grid", "modal", "grid modal"
+            surfaces="grid",  # default = "grid"
+
+            # Markdown-formatted text that describes the panel. This is
+            # rendererd in a tooltip when the help icon in the panel
+            # title is hovered over
+            help_markdown="A description of the panel",
         )
 
+The ``surfaces`` key defines the panel's scope:
+
+-   Grid panels can be accessed from the ``+`` button in the App's
+    :ref:`grid view <app-fields-sidebar>`, which allows you to build macro
+    experiences that work with entire datasets or views
+-   Modal panels can be accessed from the ``+`` button in the App's
+    :ref:`modal view <app-sample-view>`, which allows you to build interactions
+    that focus on individual samples and scenarios
+
 .. _panel-execution-context:
 
 Execution context

docs/source/release-notes.rst

@@ -3,14 +3,125 @@ FiftyOne Release Notes
 
 .. default-role:: code
 
+FiftyOne Teams 2.1.0
+--------------------
+*Released October 1, 2024*
+
+Includes all updates from :ref:`FiftyOne 1.0.0 <release-notes-v1.0.0>`, plus:
+
+- Super admins can now migrate their deployments to
+  :ref:`Internal Mode <internal-mode>` via the
+  :ref:`Super Admin UI <super-admin-ui>`
+- Added support for sending user invitations in
+  :ref:`Internal Mode <internal-mode>`
+- Optimized performance of the :ref:`dataset page <teams-homepage>`
+- Fixed a BSON serialization bug that could cause errors when cloning or
+  exporting certain dataset views from the Teams UI
+
+.. _release-notes-v1.0.0:
+
+FiftyOne 1.0.0
+--------------
+*Released October 1, 2024*
+
+What's New
+
+- The `FiftyOne Brain <https://github.com/voxel51/fiftyone-brain>`_ is now
+  fully open source. Contributions are welcome!
+- Added :ref:`Modal Panels <panel-config>`, bringing the ability to develop and
+  use panels in the App's sample modal
+  `#4625 <https://github.com/voxel51/fiftyone/pull/4625>`_
+- All datasets now have :ref:`automatically populated <default-sample-fields>`
+  `created_at` and `last_modified_at` fields on their samples and frames
+  `#4597 <https://github.com/voxel51/fiftyone/pull/4597>`_
+- Added support for loading
+  :ref:`remotely-sourced zoo datasets <dataset-zoo-remote>` whose
+  download/preparation instructions are stored in GitHub or public URLs
+  `#4752 <https://github.com/voxel51/fiftyone/pull/4752>`_
+- Added support for loading
+  :ref:`remotely-sourced zoo models <model-zoo-remote>` whose definitions are
+  stored in GitHub or public URLs
+  `#4786 <https://github.com/voxel51/fiftyone/pull/4786>`_
+- Added `Med-SAM2 <https://arxiv.org/abs/2408.00874>`_ to the
+  :ref:`model zoo <model-zoo-med-sam-2-video-torch>`!
+  `#4733 <https://github.com/voxel51/fiftyone/pull/4733>`_,
+  `#4828 <https://github.com/voxel51/fiftyone/pull/4828>`_
+
+App
+
+- Added dozens of :ref:`builtin operators <using-operators>` for performing
+  common operations directly from the App
+  `#4830 <https://github.com/voxel51/fiftyone/pull/4830>`_
+- Label overlays in the grid are now scaled proportionally to grid zoom
+  `#4747 <https://github.com/voxel51/fiftyone/pull/4747>`_
+- Improved support for visualizing and filtering |DynamicEmbeddedDocument| list
+  fields
+  `#4833 <https://github.com/voxel51/fiftyone/pull/4833>`_
+- Added a new timeline API for synchronizing playback of multiple modal panels
+  `#4772 <https://github.com/voxel51/fiftyone/pull/4772>`_
+- Improved UI, documentation, and robustness when working with
+  :ref:`custom color schemes <app-color-schemes-app>`
+  `#4763 <https://github.com/voxel51/fiftyone/pull/4763>`_
+- Fixed a bug where the active group slice was not being persisted when
+  navigating between groups in the modal
+  `#4836 <https://github.com/voxel51/fiftyone/pull/4836>`_
+- Fixed a bug when selecting samples in grouped datasets in the modal
+  `#4789 <https://github.com/voxel51/fiftyone/pull/4789>`_
+- Fixed :ref:`heatmaps <heatmaps>` rendering for values outside of the `range`
+  attribute `#4865 <https://github.com/voxel51/fiftyone/pull/4865>`_
+
+Core
+
+- Added support for creating :ref:`summary fields <summary-fields>` to optimize
+  queries on large datasets with many objects
+  `#4765 <https://github.com/voxel51/fiftyone/pull/4765>`_
+- Dataset fields now have automatically populated `created_at` attributes
+  `#4730 <https://github.com/voxel51/fiftyone/pull/4730>`_
+- Upgraded the
+  :meth:`delete_samples() <fiftyone.core.dataset.Dataset.delete_samples>`
+  and :meth:`clear_frames() <fiftyone.core.dataset.Dataset.clear_frames>`
+  methods to support bulk deletions of 100k+ samples/frames
+  `#4787 <https://github.com/voxel51/fiftyone/pull/4787>`_
+- The :meth:`default_sidebar_groups() <fiftyone.core.odm.dataset.DatasetAppConfig.default_sidebar_groups>`
+  method now correctly handles datetime fields
+  `#4815 <https://github.com/voxel51/fiftyone/pull/4815>`_
+- Fixed an off-by-one error when converting semantic segmentations to/from
+  instance segmentations
+  `#4826 <https://github.com/voxel51/fiftyone/pull/4826>`_
+- Protect against infinitely growing content size batchers
+  `#4806 <https://github.com/voxel51/fiftyone/pull/4806>`_
+- Removed the deprecated `remove_sample()` and `remove_samples()` methods from
+  the |Dataset| class
+  `#4832 <https://github.com/voxel51/fiftyone/pull/4832>`_
+- Deprecated :ref:`Python 3.8 support <deprecation-python-3.8>`
+
+Plugins
+
+- Added
+  :meth:`ctx.group_slice <fiftyone.operators.executor.ExecutionContext.group_slice>`
+  to the operator execution context
+  `#4850 <https://github.com/voxel51/fiftyone/pull/4850>`_
+- Added
+  :meth:`set_group_slice() <fiftyone.operators.operations.Operations.set_group_slice>`
+  to the operator execution context
+  `#4844 <https://github.com/voxel51/fiftyone/pull/4844>`_
+- Improved styling for :class:`GridView <fiftyone.operators.types.GridView>`
+  components
+  `#4764 <https://github.com/voxel51/fiftyone/pull/4764>`_
+- A loading error is now displayed in the actions row when operators with
+  :ref:`placements <operator-placement>` fail to load
+  `#4714 <https://github.com/voxel51/fiftyone/pull/4714>`_
+- Ensure the App loads when plugins fail to load
+  `#4769 <https://github.com/voxel51/fiftyone/pull/4769>`_
+
+.. _release-notes-v0.25.2:
 
 FiftyOne 0.25.2
 ---------------
 *Released September 19, 2024*
 
-* Require `pymongo<4.9` to fix database connections
-* Require `pydicom<3` for :ref:`DICOM datasets <DICOMDataset-import>`
-
+- Require `pymongo<4.9` to fix database connections
+- Require `pydicom<3` for :ref:`DICOM datasets <DICOMDataset-import>`
 
 FiftyOne Teams 2.0.1
 --------------------

docs/source/user_guide/using_datasets.rst

@@ -1635,16 +1635,14 @@ editable at any time:
 Summary fields
 --------------
 
-Summary fields allow you to efficiently perform queries where directly querying
-the underlying field is prohibitively slow due to the number of objects/frames
-in the field.
+Summary fields allow you to efficiently perform queries on large datasets where
+directly querying the underlying field is prohibitively slow due to the number
+of objects/frames in the field.
 
-For example, as we'll see below, summary fields are useful for retrieving
-samples in a video dataset that contain specific values of interest in at least
-one frame.
-
-Use :meth:`create_summary_field() <fiftyone.core.dataset.Dataset.create_summary_field>`
-to create a summary field for a given input field path:
+For example, suppose you're working on a
+:ref:`video dataset <video-datasets>` with frame-level objects, and you're
+interested in finding videos that contain specific classes of interest, eg
+`person`, in at least one frame:
 
 .. code-block:: python
     :linenos:
@@ -1656,6 +1654,25 @@ to create a summary field for a given input field path:
     dataset = foz.load_zoo_dataset("quickstart-video")
     dataset.set_field("frames.detections.detections.confidence", F.rand()).save()
 
+    session = fo.launch_app(dataset)
+
+.. image:: /images/datasets/quickstart-video.gif
+   :alt: quickstart-video
+   :align: center
+
+One approach is to directly query the frame-level field (`frames.detections`
+in this case) in the App's sidebar. However, when the dataset is large, such
+queries are inefficient, as they cannot
+:ref:`leverage indexes <app-indexed-filtering>` and thus require full
+collection scans over all frames to retrieve the relevant samples.
+
+A more efficient approach is to first use
+:meth:`create_summary_field() <fiftyone.core.dataset.Dataset.create_summary_field>`
+to summarize the relevant input field path(s):
+
+.. code-block:: python
+    :linenos:
+
     # Generate a summary field for object labels
     field_name = dataset.create_summary_field("frames.detections.detections.label")
 
@@ -1666,15 +1683,6 @@ to create a summary field for a given input field path:
     # Generate a summary field for [min, max] confidences
     dataset.create_summary_field("frames.detections.detections.confidence")
 
-.. note::
-
-    Summary fields are :ref:`read-only <read-only-fields>`, as they are
-    implicitly derived from the contents of their source field and are not
-    intended to be directly modified.
-
-    They are also :ref:`indexed <app-indexed-filtering>` by default, so
-    filtering them :ref:`in the App <app-indexed-filtering>` is performant.
-
 Summary fields can be generated for sample-level and frame-level fields, and
 the input fields can be either categorical or numeric:
 
@@ -1753,6 +1761,28 @@ the input fields can be either categorical or numeric:
             ]
             """
 
+As the above examples illustrate, summary fields allow you to encode various
+types of information at the sample-level that you can directly query to find
+samples that contain specific values.
+
+Moreover, summary fields are :ref:`indexed <app-indexed-filtering>` by default
+and the App can natively leverage these indexes to provide performant
+filtering:
+
+.. image:: /images/datasets/quickstart-video-summary-fields.gif
+   :alt: quickstart-video-summary-fields
+   :align: center
+
+.. note::
+
+    Summary fields are automatically added to a `summaries`
+    :ref:`sidebar group <dataset-app-config-sidebar-groups>` in the App for
+    easy access and organization.
+
+    They are also :ref:`read-only <read-only-fields>` by default, as they are
+    implicitly derived from the contents of their source field and are not
+    intended to be directly modified.
+
 You can use
 :meth:`list_summary_fields() <fiftyone.core.dataset.Dataset.list_summary_fields>`
 to list the names of the summary fields on your dataset: