cbrkit 0.23.0__tar.gz → 0.23.2__tar.gz

{cbrkit-0.23.0/src/cbrkit.egg-info → cbrkit-0.23.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cbrkit
-Version: 0.23.0
+Version: 0.23.2
 Summary: Customizable Case-Based Reasoning (CBR) toolkit for Python with a built-in API and CLI
 Author-email: Mirko Lenz <mirko@mirkolenz.com>
 Project-URL: Repository, https://github.com/wi2trier/cbrkit
@@ -35,12 +35,12 @@ Requires-Dist: pandas<3,>=2
 Requires-Dist: polars<2,>=1
 Requires-Dist: pydantic<3,>=2
 Requires-Dist: pyyaml<7,>=6
-Requires-Dist: rich<14,>=13
+Requires-Dist: rich<15,>=13
 Requires-Dist: rtoml<1,>=0.12
 Requires-Dist: scipy<2,>=1
 Requires-Dist: xmltodict<1,>=0.13
 Provides-Extra: all
-Requires-Dist: cbrkit[api,cli,eval,graphs,llm,nlp,timeseries,transformers]; extra == "all"
+Requires-Dist: cbrkit[api,cli,eval,graphs,graphviz,llm,nlp,timeseries,transformers]; extra == "all"
 Provides-Extra: api
 Requires-Dist: cbrkit[cli]; extra == "api"
 Requires-Dist: fastapi<1,>=0.100; extra == "api"
@@ -54,6 +54,8 @@ Requires-Dist: ranx<1,>=0.3; extra == "eval"
 Provides-Extra: graphs
 Requires-Dist: networkx<4,>=3; extra == "graphs"
 Requires-Dist: rustworkx<1,>=0.15; extra == "graphs"
+Provides-Extra: graphviz
+Requires-Dist: pygraphviz<2,>=1; extra == "graphviz"
 Provides-Extra: llm
 Requires-Dist: cohere<6,>=5; extra == "llm"
 Requires-Dist: voyageai<1,>=0.3; extra == "llm"
@@ -72,7 +74,7 @@ Requires-Dist: spacy<4,>=3.8; ((sys_platform == "darwin" and platform_machine ==
 Provides-Extra: timeseries
 Requires-Dist: minineedle<4,>=3; extra == "timeseries"
 Provides-Extra: transformers
-Requires-Dist: sentence-transformers<4,>=3; extra == "transformers"
+Requires-Dist: sentence-transformers<5,>=3; extra == "transformers"
 Requires-Dist: torch<2.3,>=2.2; (sys_platform == "darwin" and platform_machine == "x86_64") and extra == "transformers"
 Requires-Dist: torch<3,>=2.5; ((sys_platform == "darwin" and platform_machine == "arm64") or sys_platform == "linux") and extra == "transformers"
 Requires-Dist: transformers<5,>=4; extra == "transformers"

{cbrkit-0.23.0 → cbrkit-0.23.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "cbrkit"
-version = "0.23.0"
+version = "0.23.2"
 description = "Customizable Case-Based Reasoning (CBR) toolkit for Python with a built-in API and CLI"
 authors = [{ name = "Mirko Lenz", email = "mirko@mirkolenz.com" }]
 readme = "README.md"
@@ -43,14 +43,14 @@ dependencies = [
     "polars>=1,<2",
     "pydantic>=2,<3",
     "pyyaml>=6,<7",
-    "rich>=13,<14",
+    "rich>=13,<15",
     "rtoml>=0.12,<1",
     "scipy>=1,<2",
     "xmltodict>=0.13,<1",
 ]
 
 [project.optional-dependencies]
-all = ["cbrkit[api,cli,eval,graphs,llm,nlp,timeseries,transformers]"]
+all = ["cbrkit[api,cli,eval,graphs,graphviz,llm,nlp,timeseries,transformers]"]
 api = [
     "cbrkit[cli]",
     "fastapi>=0.100,<1",
@@ -61,6 +61,7 @@ api = [
 cli = ["typer>=0.9,<1"]
 eval = ["ranx>=0.3,<1"]
 graphs = ["networkx>=3,<4", "rustworkx>=0.15,<1"]
+graphviz = ["pygraphviz>=1,<2"]
 llm = [
     "cohere>=5,<6",
     "voyageai>=0.3,<1",
@@ -80,7 +81,7 @@ nlp = [
 ]
 timeseries = ["minineedle>=3,<4"]
 transformers = [
-    "sentence-transformers>=3,<4",
+    "sentence-transformers>=3,<5",
     "torch>=2.2,<2.3; sys_platform == 'darwin' and platform_machine == 'x86_64'",
     "torch>=2.5,<3; (sys_platform == 'darwin' and platform_machine == 'arm64') or sys_platform == 'linux'",
     "transformers>=4,<5",

{cbrkit-0.23.0 → cbrkit-0.23.2}/src/cbrkit/model/graph.py

@@ -26,6 +26,7 @@ __all__ = [
     "from_networkx",
     "to_networkx",
     "to_sequence",
+    "to_graphviz",
 ]
 
 type GraphElementType = Literal["node", "edge"]
@@ -359,3 +360,46 @@ with optional_dependencies():
         )
 
         return Graph(nodes=nodes, edges=edges, value=g.graph)
+
+
+with optional_dependencies():
+    from pygraphviz import AGraph
+
+    def to_graphviz[N, E, G](
+        g: Graph[Any, N, E, G],
+        name: str,
+        strict: bool,
+        directed: bool,
+        node_converter: ConversionFunc[N, Mapping[str, str]] | None = None,
+        edge_converter: ConversionFunc[E, Mapping[str, str]] | None = None,
+        graph_converter: ConversionFunc[G, Mapping[str, str]] | None = None,
+        node_attr: Mapping[str, str] | None = None,
+        edge_attr: Mapping[str, str] | None = None,
+        graph_attr: Mapping[str, str] | None = None,
+    ) -> AGraph:
+        gv = AGraph(
+            name=name,
+            strict=strict,
+            directed=directed,
+        )
+        gv.graph_attr.update(graph_attr or {})
+        gv.node_attr.update(node_attr or {})
+        gv.edge_attr.update(edge_attr or {})
+
+        if node_converter is not None:
+            for node in g.nodes.values():
+                gv.add_node(node.key, **node_converter(node.value))
+
+        if edge_converter is not None:
+            for edge in g.edges.values():
+                gv.add_edge(
+                    edge.source.key,
+                    edge.target.key,
+                    key=edge.key,
+                    **edge_converter(edge.value),
+                )
+
+        if graph_converter is not None:
+            gv.graph_attr.update(graph_converter(g.value))
+
+        return gv

{cbrkit-0.23.0 → cbrkit-0.23.2}/src/cbrkit/retrieval/build.py

@@ -11,6 +11,7 @@ from ..helpers import (
     get_value,
     mp_count,
     mp_map,
+    mp_starmap,
     sim_map2ranking,
     unpack_float,
     use_mp,
@@ -154,20 +155,30 @@ class build[K, V, S: Float](RetrieverFunc[K, V, S]):
     multiprocessing: Pool | int | bool = False
     chunksize: int | None = None
 
+    def __call_single__(self, x: Casebase[K, V], y: V) -> SimMap[K, S]:
+        sim_func = batchify_sim(self.similarity_func)
+        flat_batches = [(x, y) for x in x.values()]
+        flat_sims: Sequence[S] = sim_func(flat_batches)
+
+        return {key: sim for key, sim in zip(x.keys(), flat_sims, strict=True)}
+
     @override
     def __call__(
         self, batches: Sequence[tuple[Casebase[K, V], V]]
     ) -> Sequence[SimMap[K, S]]:
+        if self.chunksize is None:
+            return mp_starmap(
+                self.__call_single__, batches, self.multiprocessing, logger
+            )
+
         sim_func = batchify_sim(self.similarity_func)
-        similarities: list[dict[K, S]] = []
+        similarities: list[dict[K, S]] = [{} for _ in range(len(batches))]
 
         flat_sims: Sequence[S] = []
         flat_batches_index: list[tuple[int, K]] = []
         flat_batches: list[tuple[V, V]] = []
 
         for idx, (casebase, query) in enumerate(batches):
-            similarities.append({})
-
             for key, case in casebase.items():
                 flat_batches_index.append((idx, key))
                 flat_batches.append((case, query))
@@ -176,8 +187,8 @@ class build[K, V, S: Float](RetrieverFunc[K, V, S]):
             chunksize = self.chunksize or math.ceil(
                 len(flat_batches) / mp_count(self.multiprocessing)
             )
-            pair_chunks = list(chunkify(flat_batches, chunksize))
-            sim_chunks = mp_map(sim_func, pair_chunks, self.multiprocessing, logger)
+            batch_chunks = list(chunkify(flat_batches, chunksize))
+            sim_chunks = mp_map(sim_func, batch_chunks, self.multiprocessing, logger)
 
             for sim_chunk in sim_chunks:
                 flat_sims.extend(sim_chunk)

{cbrkit-0.23.0 → cbrkit-0.23.2/src/cbrkit.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cbrkit
-Version: 0.23.0
+Version: 0.23.2
 Summary: Customizable Case-Based Reasoning (CBR) toolkit for Python with a built-in API and CLI
 Author-email: Mirko Lenz <mirko@mirkolenz.com>
 Project-URL: Repository, https://github.com/wi2trier/cbrkit
@@ -35,12 +35,12 @@ Requires-Dist: pandas<3,>=2
 Requires-Dist: polars<2,>=1
 Requires-Dist: pydantic<3,>=2
 Requires-Dist: pyyaml<7,>=6
-Requires-Dist: rich<14,>=13
+Requires-Dist: rich<15,>=13
 Requires-Dist: rtoml<1,>=0.12
 Requires-Dist: scipy<2,>=1
 Requires-Dist: xmltodict<1,>=0.13
 Provides-Extra: all
-Requires-Dist: cbrkit[api,cli,eval,graphs,llm,nlp,timeseries,transformers]; extra == "all"
+Requires-Dist: cbrkit[api,cli,eval,graphs,graphviz,llm,nlp,timeseries,transformers]; extra == "all"
 Provides-Extra: api
 Requires-Dist: cbrkit[cli]; extra == "api"
 Requires-Dist: fastapi<1,>=0.100; extra == "api"
@@ -54,6 +54,8 @@ Requires-Dist: ranx<1,>=0.3; extra == "eval"
 Provides-Extra: graphs
 Requires-Dist: networkx<4,>=3; extra == "graphs"
 Requires-Dist: rustworkx<1,>=0.15; extra == "graphs"
+Provides-Extra: graphviz
+Requires-Dist: pygraphviz<2,>=1; extra == "graphviz"
 Provides-Extra: llm
 Requires-Dist: cohere<6,>=5; extra == "llm"
 Requires-Dist: voyageai<1,>=0.3; extra == "llm"
@@ -72,7 +74,7 @@ Requires-Dist: spacy<4,>=3.8; ((sys_platform == "darwin" and platform_machine ==
 Provides-Extra: timeseries
 Requires-Dist: minineedle<4,>=3; extra == "timeseries"
 Provides-Extra: transformers
-Requires-Dist: sentence-transformers<4,>=3; extra == "transformers"
+Requires-Dist: sentence-transformers<5,>=3; extra == "transformers"
 Requires-Dist: torch<2.3,>=2.2; (sys_platform == "darwin" and platform_machine == "x86_64") and extra == "transformers"
 Requires-Dist: torch<3,>=2.5; ((sys_platform == "darwin" and platform_machine == "arm64") or sys_platform == "linux") and extra == "transformers"
 Requires-Dist: transformers<5,>=4; extra == "transformers"

{cbrkit-0.23.0 → cbrkit-0.23.2}/src/cbrkit.egg-info/requires.txt

@@ -4,7 +4,7 @@ pandas<3,>=2
 polars<2,>=1
 pydantic<3,>=2
 pyyaml<7,>=6
-rich<14,>=13
+rich<15,>=13
 rtoml<1,>=0.12
 scipy<2,>=1
 xmltodict<1,>=0.13
@@ -16,7 +16,7 @@ numpy<3,>=2
 numpy<2,>=1
 
 [all]
-cbrkit[api,cli,eval,graphs,llm,nlp,timeseries,transformers]
+cbrkit[api,cli,eval,graphs,graphviz,llm,nlp,timeseries,transformers]
 
 [api]
 cbrkit[cli]
@@ -35,6 +35,9 @@ ranx<1,>=0.3
 networkx<4,>=3
 rustworkx<1,>=0.15
 
+[graphviz]
+pygraphviz<2,>=1
+
 [llm]
 cohere<6,>=5
 voyageai<1,>=0.3
@@ -60,7 +63,7 @@ spacy<3.8,>=3.7
 minineedle<4,>=3
 
 [transformers]
-sentence-transformers<4,>=3
+sentence-transformers<5,>=3
 transformers<5,>=4
 
 [transformers:(sys_platform == "darwin" and platform_machine == "arm64") or sys_platform == "linux"]

{cbrkit-0.23.0 → cbrkit-0.23.2}/tests/test_retrieve.py

@@ -32,6 +32,7 @@ def test_retrieve_multiprocessing():
                 aggregator=cbrkit.sim.aggregator(pooling="mean"),
             ),
             multiprocessing=2,
+            chunksize=5,
         )
     )
     result = cbrkit.retrieval.apply_query(