Prepare phy 2.1.0 release

Author: rossant

.release-smoke/latest-testpypi-version.txt

@@ -2,36 +2,29 @@
 
 [![Install and Test with Pip](https://github.com/cortex-lab/phy/actions/workflows/python-test.yml/badge.svg)](https://github.com/cortex-lab/phy/actions/workflows/python-test.yml)
 [![codecov.io](https://img.shields.io/codecov/c/github/cortex-lab/phy.svg)](http://codecov.io/github/cortex-lab/phy)
-[![Documentation](https://img.shields.io/badge/docs-phy.cortexlab.net-blue.svg)](https://phy.cortexlab.net)
+[![Documentation](https://img.shields.io/badge/docs-Read_the_Docs-blue.svg)](https://phy.readthedocs.io/en/latest/)
 [![GitHub release](https://img.shields.io/github/release/cortex-lab/phy.svg)](https://github.com/cortex-lab/phy/releases/latest)
 [![PyPI release](https://img.shields.io/pypi/v/phy.svg)](https://pypi.python.org/pypi/phy)
 
 [**phy**](https://github.com/cortex-lab/phy) is an open-source Python library providing a graphical user interface for visualization and manual curation of large-scale electrophysiological data. It is optimized for high-density multielectrode arrays containing hundreds to thousands of recording sites, especially Neuropixels recordings.
 
-> **Release candidate available:** `phy 2.1.0rc1` is available for testing on PyPI. This maintenance-focused release candidate aims to improve installation and GUI reliability on current systems. See the [release notes](https://phy.readthedocs.io/en/latest/release/) for installation instructions, compatibility notes, and testing guidance.
+> **Current release:** `phy 2.1.0` is a maintenance-focused release that improves installation and GUI reliability on current systems. See the [release notes](https://phy.readthedocs.io/en/latest/release/) for details and compatibility notes.
 
-[![phy 2.1.0rc1 screenshot](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)
+[![phy 2.1.0 screenshot](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)
 
 ## Current status
 
-As of March 2026, `phy 2.1.0rc1` is a maintenance-focused release candidate for the current 2.x line.
+As of July 2026, `phy 2.1.0` is the current stable release for the 2.x line.
 
 The main goals of this release are:
 
 * dependency and packaging modernization
 * replacing a fragile legacy web-based GUI component with a Qt-native implementation
 * improving display reliability on modern systems
-* collecting feedback from beta testers before a final `2.1.0` release
+* incorporating fixes identified during release-candidate testing
 
 Dataset formats are unchanged. Some plugins that relied on internal HTML or web-based GUI components may need updates.
 
-If you would like to test this release candidate, install it in a fresh Python 3.10+ environment:
-
-```bash
-python -m pip install --upgrade pip
-pip install "phy==2.1.0rc1"
-```
-
 Please report any issues or compatibility regressions on [GitHub issues](https://github.com/cortex-lab/phy/issues).
 
 ## Supported workflows
@@ -42,7 +35,7 @@ phy currently provides three main entry points:
 * **Kwik GUI**: a legacy workflow for datasets sorted with klusta and klustakwik2
 * **Trace GUI**: an experimental raw-data viewer for opening continuous electrophysiology recordings directly
 
-Current testing and maintenance work is focused on modern Linux, macOS, and Windows environments. Linux is still the best-covered platform, but cross-platform testing is active during the `2.1.0rc1` cycle.
+Current testing and maintenance work is focused on modern Linux, macOS, and Windows environments. Linux is still the best-covered platform.
 
 ## Installation
 
@@ -61,12 +54,6 @@ If you plan to use the legacy Kwik GUI, also install:
 pip install klusta klustakwik2
 ```
 
-For release-candidate testing specifically, install the exact RC version instead:
-
-```bash
-pip install "phy==2.1.0rc1"
-```
-
 ## Quick start
 
 Open the Template GUI on a spike sorting output directory containing `params.py`:

README.md

@@ -1,6 +1,6 @@
-# phy 2.1.0rc1 announcement text
+# phy 2.1.0 announcement text
 
-Short copy-paste announcement text for the `2.1.0rc1` release candidate.
+Short copy-paste announcement text for the final `2.1.0` release.
 
 For installation, upgrade, compatibility, and testing details, point people to the live release notes page:
 https://phy.readthedocs.io/en/latest/release/
@@ -10,19 +10,19 @@ https://phy.readthedocs.io/en/latest/release/
 Subject:
 
 ```text
-phy 2.1.0rc1 release candidate available for testing
+phy 2.1.0 released
 ```
 
 Body:
 
 ```text
 Dear phy users,
 
-phy 2.1.0rc1 is now available on PyPI for testing. This is a release candidate for a maintenance release of the current 2.x line, focused on fixing several long-standing issues that have built up over recent years while the project had very limited maintenance.
+phy 2.1.0 is now available on PyPI. This maintenance release of the current 2.x line focuses on fixing several long-standing issues that built up over recent years while the project had very limited maintenance.
 
 This release does not aim to introduce major new features. The goal is to make phy easier to install and more reliable again on current systems, especially for users who have run into packaging or GUI problems.
 
-If you still use phy, testing and feedback would be very helpful before the final 2.1.0 release.
+Thank you to everyone who tested the release candidate and reported issues.
 
 Release notes and compatibility notes:
 https://phy.readthedocs.io/en/latest/release/
@@ -34,12 +34,12 @@ https://github.com/cortex-lab/phy/issues
 ## Copy-paste announcement for GitHub release notes
 
 ```text
-phy 2.1.0rc1 is now available on PyPI for testing.
+phy 2.1.0 is now available on PyPI.
 
-This is a release candidate for a maintenance release of the current 2.x line, focused on fixing several long-standing issues that have accumulated over recent years.
+This maintenance release of the current 2.x line focuses on fixing several long-standing issues that accumulated over recent years.
 
 The goal of this release is not to add major new features, but to make phy easier to install and more reliable again on current systems.
 
-For release notes, compatibility notes, and testing guidance, see:
+For release notes and compatibility notes, see:
 https://phy.readthedocs.io/en/latest/release/
 ```

docs/announcement_2.1.0rc1.md docs/announcement_2.1.0.mddocs/announcement_2.1.0rc1.md renamed to docs/announcement_2.1.0.md

@@ -2,9 +2,9 @@
 
 [**phy**](https://github.com/cortex-lab/phy) is an open-source Python library providing a graphical user interface for visualization and manual curation of large-scale electrophysiological data. It is optimized for high-density multielectrode arrays containing hundreds to thousands of recording sites (mostly [Neuropixels probes](https://www.ucl.ac.uk/neuropixels/)).
 
-> **Current release candidate:** [phy 2.1.0rc1](release.md) is available for testing on PyPI. This maintenance-focused release candidate aims to improve installation and GUI reliability on current systems. See the [release notes](release.md) for installation instructions, compatibility notes, and testing guidance.
+> **Current release:** [phy 2.1.0](release.md) is a maintenance-focused release that improves installation and GUI reliability on current systems. See the [release notes](release.md) for details and compatibility notes.
 
-[![phy 2.1.0rc1 screenshot](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)
+[![phy 2.1.0 screenshot](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)](https://user-images.githubusercontent.com/1942359/74028054-c284b880-49a9-11ea-8815-1b7e727a8644.png)
 
 ## Spike sorting programs
 
@@ -20,4 +20,4 @@ klusta is a legacy spike sorting program based on an automatic clustering algori
 
 ## Installation instructions
 
-See the [GitHub repository](https://github.com/cortex-lab/phy/) for stable-install instructions, or go to the [release notes](release.md) for the `2.1.0rc1` release-candidate install and testing instructions.
+See the [GitHub repository](https://github.com/cortex-lab/phy/) for installation instructions, or go to the [release notes](release.md) for details about `2.1.0`.

docs/index.md

@@ -1,13 +1,20 @@
 # Release notes
 
-Current version is phy v2.1.0rc1 (release candidate 1) (11 Mar 2026).
+Current version is phy v2.1.0 (17 Jul 2026).
 
-## phy 2.1.0 (upcoming)
+## phy 2.1.0 (17 Jul 2026)
+
+This final release incorporates the fixes identified during the `2.1.0rc1` testing period.
 
 ### Added
 
 * Code of Conduct for project participation and incident reporting.
 
+### Fixed
+
+* Preserve the ClusterView sort order when cluster metadata changes ([#1375](https://github.com/cortex-lab/phy/issues/1375)).
+* Display NumPy-typed values correctly in native Qt ClusterView columns ([#1377](https://github.com/cortex-lab/phy/issues/1377)).
+
 
 ## phy 2.1.0rc1
 
@@ -105,8 +112,7 @@ When reporting issues, please include:
 
 ## Historical notes for older releases
 
-The notes below are kept for reference only and concern older releases, not the
-current `2.1.0rc1` release candidate.
+The notes below are kept for reference only and concern older releases.
 
 ### phy 2.0 beta 1 (2020-02)
 

docs/release.md

@@ -6,7 +6,7 @@ Here are a few tips if something goes wrong.
 
 Look at existing [GitHub issues](https://github.com/cortex-lab/phy/issues) to see if someone else had the same problem. If not, feel free to open a new issue by giving a descriptive title and a comprehensive description of the problem, including screenshots if needed.
 
-If you are testing `phy 2.1.0rc1`, please also include:
+If you are reporting a problem with `phy 2.1.0`, please also include:
 
 * operating system
 * Python version

docs/troubleshooting.md

@@ -27,7 +27,7 @@
 
 __author__ = 'Cyrille Rossant'
 __email__ = 'cyrille.rossant at gmail.com'
-__version__ = '2.1.0rc1'
+__version__ = '2.1.0'
 __version_git__ = __version__ + _git_version()
 
 

phy/__init__.py

@@ -190,9 +190,7 @@ def get_mean_spike_raw_amplitudes(self, cluster_id):
         spike_ids = self._get_amplitude_spike_ids(cluster_id)
         return np.mean(self.get_spike_raw_amplitudes(spike_ids))
 
-    def _get_waveforms_with_n_spikes(
-        self, cluster_id, n_spikes_waveforms, current_filter=None
-    ):
+    def _get_waveforms_with_n_spikes(self, cluster_id, n_spikes_waveforms, current_filter=None):
         # HACK: we pass self.raw_data_filter.current_filter so that it is cached properly.
         pos = self.model.channel_positions
 
@@ -205,9 +203,7 @@ def _get_waveforms_with_n_spikes(
         # Or keep spikes from a subset of the chunks for performance reasons (decompression will
         # happen on the fly here).
         else:
-            spike_ids = self.selector(
-                n_spikes_waveforms, [cluster_id], subset_chunks=True
-            )
+            spike_ids = self.selector(n_spikes_waveforms, [cluster_id], subset_chunks=True)
 
         # Get the best channels.
         channel_ids = self.get_best_channels(cluster_id)
@@ -250,9 +246,7 @@ def _set_view_creator(self):
         self.view_creator['WaveformView'] = self.create_waveform_view
 
     def _get_waveforms_dict(self):
-        waveform_functions = _concatenate_parents_attributes(
-            self.__class__, '_waveform_functions'
-        )
+        waveform_functions = _concatenate_parents_attributes(self.__class__, '_waveform_functions')
         return {name: getattr(self, method) for name, method in waveform_functions}
 
     def create_waveform_view(self):
@@ -378,9 +372,7 @@ def _get_feature_view_spike_times(self, cluster_id=None, load_all=False):
         if len(spike_ids) == 0:
             return
         spike_times = self._get_spike_times_reordered(spike_ids)
-        return Bunch(
-            data=spike_times, spike_ids=spike_ids, lim=(0.0, self.model.duration)
-        )
+        return Bunch(data=spike_times, spike_ids=spike_ids, lim=(0.0, self.model.duration))
 
     def _get_spike_features(self, spike_ids, channel_ids):
         if len(spike_ids) == 0:  # pragma: no cover
@@ -412,10 +404,7 @@ def _get_features(self, cluster_id=None, channel_ids=None, load_all=False):
         return self._get_spike_features(spike_ids, channel_ids)
 
     def create_feature_view(self):
-        if (
-            self.model.features is None
-            and getattr(self.model, 'spike_waveforms', None) is None
-        ):
+        if self.model.features is None and getattr(self.model, 'spike_waveforms', None) is None:
             # NOTE: we can still construct the feature view when there are spike waveforms.
             return
         view = FeatureView(
@@ -579,9 +568,7 @@ def _get_template_waveforms(self, cluster_id):
         masks = count / float(count.max())
         masks = np.tile(masks.reshape((-1, 1)), (1, len(channel_ids)))
         # Get all templates from which this cluster stems from.
-        templates = [
-            self.model.get_template(template_id) for template_id in template_ids
-        ]
+        templates = [self.model.get_template(template_id) for template_id in template_ids]
         # Construct the waveforms array.
         ns = self.model.n_samples_waveforms
         data = np.zeros((len(template_ids), ns, self.model.n_channels))
@@ -922,9 +909,7 @@ def __init__(
         _add_log_file(Path(dir_path) / 'phy.log')
 
         # Create or reuse a Model instance (any object)
-        self.model = (
-            self._create_model(dir_path=dir_path, **kwargs) if model is None else model
-        )
+        self.model = self._create_model(dir_path=dir_path, **kwargs) if model is None else model
 
         # Set up the cache.
         self._set_cache(clear_cache)
@@ -945,19 +930,15 @@ def __init__(
         # The controller.default_views can be set by the child class, otherwise it is computed
         # by concatenating all parents _new_views.
         if getattr(self, 'default_views', None) is None:
-            self.default_views = _concatenate_parents_attributes(
-                self.__class__, '_new_views'
-            )
+            self.default_views = _concatenate_parents_attributes(self.__class__, '_new_views')
         self._async_callers = {}
         self.config_dir = config_dir
 
         # Clear the GUI state files if needed.
         if clear_state:
             self._clear_state()
 
-        self.selection = Selection(
-            self
-        )  # keep track of selected clusters, spikes, channels, etc.
+        self.selection = Selection(self)  # keep track of selected clusters, spikes, channels, etc.
 
         # Attach plugins before setting up the supervisor, so that plugins
         # can register callbacks to events raised during setup.
@@ -1021,9 +1002,7 @@ def _set_view_creator(self):
             'ClusterScatterView': self.create_cluster_scatter_view,
             'CorrelogramView': self.create_correlogram_view,
             'ISIView': self._make_histogram_view(ISIView, self._get_isi),
-            'FiringRateView': self._make_histogram_view(
-                FiringRateView, self._get_firing_rate
-            ),
+            'FiringRateView': self._make_histogram_view(FiringRateView, self._get_firing_rate),
             'AmplitudeView': self.create_amplitude_view,
             'ProbeView': self.create_probe_view,
             'RasterView': self.create_raster_view,
@@ -1032,9 +1011,7 @@ def _set_view_creator(self):
         # Spike attributes.
         for name, arr in getattr(self.model, 'spike_attributes', {}).items():
             view_name = f'Spike{name.title()}View'
-            self.view_creator[view_name] = self._make_spike_attributes_view(
-                view_name, name, arr
-            )
+            self.view_creator[view_name] = self._make_spike_attributes_view(view_name, name, arr)
 
     def _set_cluster_metrics(self):
         """Set the cluster metrics dictionary with some default metrics."""
@@ -1375,9 +1352,7 @@ def _amplitude_getter(self, cluster_ids, name=None, load_all=False):
         out = []
         n = self.n_spikes_amplitudes if not load_all else None
         # Find the first cluster, used to determine the best channels.
-        first_cluster = next(
-            cluster_id for cluster_id in cluster_ids if cluster_id is not None
-        )
+        first_cluster = next(cluster_id for cluster_id in cluster_ids if cluster_id is not None)
         # Best channels of the first cluster.
         channel_ids = self.get_best_channels(first_cluster)
         # Best channel of the first cluster.
@@ -1694,9 +1669,7 @@ def create_misc_actions(self, gui):
         )
         def toggle_spike_reorder(checked):
             """Toggle spike time reordering."""
-            logger.debug(
-                '%s spike time reordering.', 'Enable' if checked else 'Disable'
-            )
+            logger.debug('%s spike time reordering.', 'Enable' if checked else 'Disable')
             emit('toggle_spike_reorder', self, checked)
 
         # Action to switch the raw data filter in the trace and waveform views.
@@ -1713,9 +1686,7 @@ def switch_raw_data_filter():
             # Update the waveform view.
             for v in gui.list_views(WaveformView):
                 if v.auto_update:
-                    v.on_select_threaded(
-                        self.supervisor, self.supervisor.selected, gui=gui
-                    )
+                    v.on_select_threaded(self.supervisor, self.supervisor.selected, gui=gui)
                     v.ex_status = filter_name
                     v.update_status()
 
@@ -1734,9 +1705,7 @@ def _add_default_color_schemes(self, view):
 
         def group_index(cluster_id):
             group = self.supervisor.cluster_meta.get('group', cluster_id)
-            return group_colors.get(
-                group, 0
-            )  # TODO: better handling of colors for custom groups
+            return group_colors.get(group, 0)  # TODO: better handling of colors for custom groups
 
         depth = self.supervisor.cluster_metrics['depth']
         fr = self.supervisor.cluster_metrics['fr']
@@ -1777,8 +1746,7 @@ def create_gui(self, default_views=None, **kwargs):
             subtitle=str(self.dir_path),
             config_dir=self.config_dir,
             local_path=self.cache_dir / 'state.json',
-            default_state_path=Path(inspect.getfile(self.__class__)).parent
-            / 'static/state.json',
+            default_state_path=Path(inspect.getfile(self.__class__)).parent / 'static/state.json',
             view_creator=self.view_creator,
             default_views=default_views,
             enable_threading=self._enable_threading,