bblean 0.6.0b2__tar.gz → 0.6.1b0__tar.gz

{bblean-0.6.0b2 → bblean-0.6.1b0}/.github/workflows/upload-to-pypi.yaml

@@ -15,10 +15,12 @@ on:
         required: false
         default: false
         type: boolean
+  release:
+    types: [published]
 
 env:
   PYTHON_VERSION: '3.11'
-  SETUPTOOLS_SCM_PRETEND_VERSION: ${{ github.event.inputs.version }}
+  SETUPTOOLS_SCM_PRETEND_VERSION: ${{ github.event_name == 'release' && github.event.release.tag_name || github.event.inputs.version }}
   # cibuildwheel configuration:
   # Skip py 3.14, 32 bit and musllinux (Alpine) wheels
   CIBW_SKIP: "cp314-* cp314t-* *-manylinux_i686 *-win32 *-musllinux_*"
@@ -93,7 +95,7 @@ jobs:
   publish_to_testpypi:
     needs: [build_wheels, make_sdist]
     runs-on: ubuntu-latest
-    if: ${{ github.event.inputs.upload-testpypi == 'true' }}
+    if: ${{ github.event_name != 'release' && github.event.inputs.upload-testpypi }}
     environment:
       name: testpypi
       url: https://test.pypi.org/p/bblean
@@ -115,7 +117,7 @@ jobs:
   publish_to_pypi:
     needs: [build_wheels, make_sdist]
     runs-on: ubuntu-latest
-    if: ${{ github.event.inputs.upload-pypi == 'true' }}
+    if: ${{ github.event_name == 'release' || github.event.inputs.upload-pypi }}
     environment:
       name: pypi
       url: https://pypi.org/p/bblean

{bblean-0.6.0b2 → bblean-0.6.1b0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bblean
-Version: 0.6.0b2
+Version: 0.6.1b0
 Summary: BitBirch-Lean Python package
 Author: The Miranda-Quintana Lab and other BitBirch developers
 Author-email: Ramon Alain Miranda Quintana <quintana@chem.ufl.edu>, Krisztina Zsigmond <kzsigmond@ufl.edu>, Ignacio Pickering <ipickering@ufl.edu>, Kenneth Lopez Perez <klopezperez@chem.ufl.edu>, Miroslav Lzicar <miroslav.lzicar@deepmedchem.com>
@@ -90,6 +90,7 @@ macOS via pip, which automatically includes C++ extensions:
 
 ```bash
 pip install bblean
+# Alternatively you can use 'uv pip install'
 bb --help
 ```
 

{bblean-0.6.0b2 → bblean-0.6.1b0}/README.md

@@ -47,6 +47,7 @@ macOS via pip, which automatically includes C++ extensions:
 
 ```bash
 pip install bblean
+# Alternatively you can use 'uv pip install'
 bb --help
 ```
 

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/_legacy/bb_int64.py

@@ -633,6 +633,7 @@ class BitBirch:
             X = X[:max_fps]
         threshold = self.threshold
         branching_factor = self.branching_factor
+
         n_features = _validate_n_features(X, input_is_packed, n_features)
         d_type = X.dtype
 
@@ -718,6 +719,7 @@ class BitBirch:
         """
         threshold = self.threshold
         branching_factor = self.branching_factor
+
         n_features = _validate_n_features(X, input_is_packed, n_features)
         d_type = X.dtype
 

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/_version.py

@@ -28,7 +28,7 @@ version_tuple: VERSION_TUPLE
 commit_id: COMMIT_ID
 __commit_id__: COMMIT_ID
 
-__version__ = version = '0.6.0b2'
-__version_tuple__ = version_tuple = (0, 6, 0, 'b2')
+__version__ = version = '0.6.1b0'
+__version_tuple__ = version_tuple = (0, 6, 1, 'b0')
 
 __commit_id__ = commit_id = None

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/bitbirch.py

@@ -47,6 +47,8 @@
 # ./LICENSES/GPL-3.0-only.txt.  If not, see <http://www.gnu.org/licenses/gpl-3.0.html>.
 r"""BitBirch 'Lean' class for fast, memory-efficient O(N) clustering"""
 from __future__ import annotations  # Stringize type annotations for no runtime overhead
+import pickle
+import sys
 import typing_extensions as tpx
 import os
 import random
@@ -1316,6 +1318,40 @@ class BitBirch:
             parts.append(f"tolerance={self.tolerance}")
         return f"{self.__class__.__name__}({', '.join(parts)})"
 
+    def save(self, path: Path | str) -> None:
+        r""":meta private:"""
+        # TODO: BitBIRCH is highly recursive. pickling may crash python,
+        # an alternative solution would be better
+        msg = (
+            "Saving large BitBIRCH trees may result in large memory peaks."
+            " An alternative serialization method may be implemented in the future"
+        )
+        warnings.warn(msg)
+        _old_limit = sys.getrecursionlimit()
+        sys.setrecursionlimit(1_000_000_000)
+        with open(path, mode="wb") as f:
+            pickle.dump(self, f)
+        sys.setrecursionlimit(_old_limit)
+
+    @classmethod
+    def load(cls, path: Path | str) -> tpx.Self:
+        r""":meta private:"""
+        # TODO: BitBIRCH is highly recursive. pickling may crash python,
+        # an alternative solution would be better
+        msg = (
+            "Loading large BitBIRCH trees may result in large memory peaks."
+            " An alternative serialization method may be implemented in the future"
+        )
+        warnings.warn(msg)
+        _old_limit = sys.getrecursionlimit()
+        sys.setrecursionlimit(1_000_000_000)
+        with open(path, mode="rb") as f:
+            tree = pickle.load(f)
+        sys.setrecursionlimit(_old_limit)
+        if not isinstance(tree, cls):
+            raise ValueError("Path does not contain a bitbirch object")
+        return tree
+
     def global_clustering(
         self,
         n_clusters: int,

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/cli.py

@@ -1096,26 +1096,29 @@ def _run(
 
     timer.end_timing("total", console, indent=False)
     console.print_peak_mem(out_dir, indent=False)
+    if save_tree:
+        if variant != "lean":
+            console.print("Can't save tree for non-lean variants", style="red")
+        else:
+            # TODO: Find alternative solution
+            tree.save_pickle(out_dir / "bitbirch.pkl")
     if variant == "lean":
-        if save_tree:
-            # TODO: BitBIRCH is highly recursive. pickling may crash python,
-            # an alternative solution would be better
-            _old_limit = sys.getrecursionlimit()
-            sys.setrecursionlimit(100_000)
-            with open(out_dir / "bitbirch.pkl", mode="wb") as f:
-                pickle.dump(tree, f)
-            sys.setrecursionlimit(_old_limit)
         tree.delete_internal_nodes()
-        # Dump outputs (peak memory, timings, config, cluster ids)
-        if save_centroids:
+    # Dump outputs (peak memory, timings, config, cluster ids)
+    if save_centroids:
+        if variant != "lean":
+            console.print("Can't save centroids for non-lean variants", style="red")
+            with open(out_dir / "clusters.pkl", mode="wb") as f:
+                pickle.dump(tree.get_cluster_mol_ids(), f)
+        else:
             output = tree.get_centroids_mol_ids()
             with open(out_dir / "clusters.pkl", mode="wb") as f:
                 pickle.dump(output["mol_ids"], f)
             with open(out_dir / "cluster-centroids-packed.pkl", mode="wb") as f:
                 pickle.dump(output["centroids"], f)
-        else:
-            with open(out_dir / "clusters.pkl", mode="wb") as f:
-                pickle.dump(tree.get_cluster_mol_ids(), f)
+    else:
+        with open(out_dir / "clusters.pkl", mode="wb") as f:
+            pickle.dump(tree.get_cluster_mol_ids(), f)
 
     collect_system_specs_and_dump_config(ctx.params)
     timer.dump(out_dir / "timings.json")

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/multiround.py

@@ -298,13 +298,8 @@ class _FinalTreeMergingRound(_TreeMergingRound):
 
         # Save clusters and exit
         if self.save_tree:
-            # TODO: BitBIRCH is highly recursive. pickling may crash python,
-            # an alternative solution would be better
-            _old_limit = sys.getrecursionlimit()
-            sys.setrecursionlimit(100_000)
-            with open(self.out_dir / "bitbirch.pkl", mode="wb") as f:
-                pickle.dump(tree, f)
-            sys.setrecursionlimit(_old_limit)
+            # TODO: Find alternative solution
+            tree.save_pickle(self.out_dir / "bitbirch.pkl")
         tree.delete_internal_nodes()
         if self.save_centroids:
             output = tree.get_centroids_mol_ids()

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/plotting.py

@@ -399,13 +399,17 @@ def dump_mol_images(
     clusters: list[list[int]],
     cluster_idx: int = 0,
     batch_size: int = 30,
+    limit: int = -1,
 ) -> None:
     r"""Dump smiles associated with a specific cluster as ``*.png`` image files"""
     if isinstance(smiles, str):
         smiles = [smiles]
     smiles = np.asarray(smiles)
     idxs = clusters[cluster_idx]
+    num = 0
     for i, idx_seq in enumerate(batched(idxs, batch_size)):
+        if num + len(idx_seq) > limit:
+            idx_seq = idx_seq[: num + len(idx_seq) - limit]
         mols = []
         for smi in smiles[list(idx_seq)]:
             mol = Chem.MolFromSmiles(smi)
@@ -415,6 +419,9 @@ def dump_mol_images(
         img = Draw.MolsToGridImage(mols, molsPerRow=5)
         with open(f"cluster_{cluster_idx}_{i}.png", "wb") as f:
             f.write(img.data)
+        num += len(idx_seq)
+        if num >= limit:
+            break
 
 
 # For internal use, dispatches a visualization workflow and optionally saves

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean/sklearn.py

@@ -131,8 +131,7 @@ class BitBirch(
             .astype(np.uint8, copy=False)
             .view(np.bool)
         )
-        # TODO: Even when both inputs are bool, this function warns for some reason
-        # I believe this may be a sklearn bug
+        # TODO: Due to a sklearn bug this performs unnecessary casts
         centers = self.subcluster_centers_.astype(np.uint8, copy=False).view(np.bool)
         argmin = pairwise_distances_argmin(X, centers, metric="jaccard")
         return self.subcluster_labels_[argmin]

{bblean-0.6.0b2 → bblean-0.6.1b0}/bblean.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bblean
-Version: 0.6.0b2
+Version: 0.6.1b0
 Summary: BitBirch-Lean Python package
 Author: The Miranda-Quintana Lab and other BitBirch developers
 Author-email: Ramon Alain Miranda Quintana <quintana@chem.ufl.edu>, Krisztina Zsigmond <kzsigmond@ufl.edu>, Ignacio Pickering <ipickering@ufl.edu>, Kenneth Lopez Perez <klopezperez@chem.ufl.edu>, Miroslav Lzicar <miroslav.lzicar@deepmedchem.com>
@@ -90,6 +90,7 @@ macOS via pip, which automatically includes C++ extensions:
 
 ```bash
 pip install bblean
+# Alternatively you can use 'uv pip install'
 bb --help
 ```