Skip to content

Commit 19e5830

Browse files
FlorentinDFlorentinD
committed
Map filtered KNN
1 parent 45a6d56 commit 19e5830

File tree

7 files changed

+1073
-2
lines changed

7 files changed

+1073
-2
lines changed

‎graphdatascience/procedure_surface/api/similarity/knn_filtered_endpoints.py‎

Lines changed: 215 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,215 @@
1+
from __future__ import annotations
2+
3+
from abc import ABC, abstractmethod
4+
from typing import Any
5+
6+
from pandas import DataFrame
7+
8+
from graphdatascience.procedure_surface.api.catalog.graph_api import GraphV2
9+
from graphdatascience.procedure_surface.api.estimation_result import EstimationResult
10+
from graphdatascience.procedure_surface.api.similarity.knn_endpoints import (
11+
KnnMutateResult,
12+
KnnStatsResult,
13+
KnnWriteResult,
14+
)
15+
16+
17+
class KnnFilteredEndpoints(ABC):
18+
"""Base class for Filtered K-Nearest Neighbors endpoints."""
19+
20+
@abstractmethod
21+
def mutate(
22+
self,
23+
G: GraphV2,
24+
mutate_relationship_type: str,
25+
mutate_property: str,
26+
node_properties: str | list[str] | dict[str, str],
27+
source_node_filter: str,
28+
target_node_filter: str,
29+
seed_target_nodes: bool | None = None,
30+
top_k: int | None = None,
31+
similarity_cutoff: float | None = None,
32+
delta_threshold: float | None = None,
33+
max_iterations: int | None = None,
34+
sample_rate: float | None = None,
35+
perturbation_rate: float | None = None,
36+
random_joins: int | None = None,
37+
random_seed: int | None = None,
38+
initial_sampler: Any | None = None,
39+
relationship_types: list[str] | None = None,
40+
node_labels: list[str] | None = None,
41+
sudo: bool | None = None,
42+
log_progress: bool = True,
43+
username: str | None = None,
44+
concurrency: Any | None = None,
45+
job_id: Any | None = None,
46+
) -> KnnMutateResult:
47+
"""Run filtered K-Nearest Neighbors in mutate mode."""
48+
...
49+
50+
@abstractmethod
51+
def stats(
52+
self,
53+
G: GraphV2,
54+
node_properties: str | list[str] | dict[str, str],
55+
source_node_filter: str,
56+
target_node_filter: str,
57+
seed_target_nodes: bool | None = None,
58+
top_k: int | None = None,
59+
similarity_cutoff: float | None = None,
60+
delta_threshold: float | None = None,
61+
max_iterations: int | None = None,
62+
sample_rate: float | None = None,
63+
perturbation_rate: float | None = None,
64+
random_joins: int | None = None,
65+
random_seed: int | None = None,
66+
initial_sampler: Any | None = None,
67+
relationship_types: list[str] | None = None,
68+
node_labels: list[str] | None = None,
69+
sudo: bool | None = None,
70+
log_progress: bool = True,
71+
username: str | None = None,
72+
concurrency: Any | None = None,
73+
job_id: Any | None = None,
74+
) -> KnnStatsResult:
75+
"""Run filtered K-Nearest Neighbors in stats mode."""
76+
...
77+
78+
@abstractmethod
79+
def stream(
80+
self,
81+
G: GraphV2,
82+
node_properties: str | list[str] | dict[str, str],
83+
source_node_filter: str,
84+
target_node_filter: str,
85+
seed_target_nodes: bool | None = None,
86+
top_k: int | None = None,
87+
similarity_cutoff: float | None = None,
88+
delta_threshold: float | None = None,
89+
max_iterations: int | None = None,
90+
sample_rate: float | None = None,
91+
perturbation_rate: float | None = None,
92+
random_joins: int | None = None,
93+
random_seed: int | None = None,
94+
initial_sampler: Any | None = None,
95+
relationship_types: list[str] | None = None,
96+
node_labels: list[str] | None = None,
97+
sudo: bool | None = None,
98+
log_progress: bool = True,
99+
username: str | None = None,
100+
concurrency: Any | None = None,
101+
job_id: Any | None = None,
102+
) -> DataFrame:
103+
"""Run filtered K-Nearest Neighbors in stream mode."""
104+
...
105+
106+
@abstractmethod
107+
def write(
108+
self,
109+
G: GraphV2,
110+
write_relationship_type: str,
111+
write_property: str,
112+
node_properties: str | list[str] | dict[str, str],
113+
source_node_filter: str,
114+
target_node_filter: str,
115+
seed_target_nodes: bool | None = None,
116+
top_k: int | None = None,
117+
similarity_cutoff: float | None = None,
118+
delta_threshold: float | None = None,
119+
max_iterations: int | None = None,
120+
sample_rate: float | None = None,
121+
perturbation_rate: float | None = None,
122+
random_joins: int | None = None,
123+
random_seed: int | None = None,
124+
initial_sampler: Any | None = None,
125+
relationship_types: list[str] | None = None,
126+
node_labels: list[str] | None = None,
127+
write_concurrency: int | None = None,
128+
write_to_result_store: bool | None = None,
129+
sudo: bool | None = None,
130+
log_progress: bool = True,
131+
username: str | None = None,
132+
concurrency: Any | None = None,
133+
job_id: Any | None = None,
134+
) -> KnnWriteResult:
135+
"""Run filtered K-Nearest Neighbors in write mode."""
136+
...
137+
138+
@abstractmethod
139+
def estimate(
140+
self,
141+
G: GraphV2 | dict[str, Any],
142+
node_properties: str | list[str] | dict[str, str],
143+
source_node_filter: str,
144+
target_node_filter: str,
145+
seed_target_nodes: bool | None = None,
146+
top_k: int | None = None,
147+
similarity_cutoff: float | None = None,
148+
delta_threshold: float | None = None,
149+
max_iterations: int | None = None,
150+
sample_rate: float | None = None,
151+
perturbation_rate: float | None = None,
152+
random_joins: int | None = None,
153+
random_seed: int | None = None,
154+
initial_sampler: Any | None = None,
155+
relationship_types: list[str] | None = None,
156+
node_labels: list[str] | None = None,
157+
sudo: bool | None = None,
158+
username: str | None = None,
159+
concurrency: Any | None = None,
160+
) -> EstimationResult:
161+
"""Estimate filtered K-Nearest Neighbors execution requirements.
162+
163+
Parameters
164+
----------
165+
G : GraphV2 | dict[str, Any]
166+
The graph to run the algorithm on.
167+
node_properties : str | list[str]
168+
The node properties to use for similarity computation.
169+
mutate_property : str
170+
The relationship property to store the similarity score in.
171+
mutate_relationship_type : str
172+
The relationship type to use for the new relationships.
173+
source_node_filter : str | None, default=None
174+
A Cypher expression to filter which nodes can be sources in the similarity computation.
175+
target_node_filter : str | None, default=None
176+
A Cypher expression to filter which nodes can be targets in the similarity computation.
177+
seed_target_nodes : bool | None, default=None
178+
Whether to use a seeded approach for target node selection.
179+
similarity_cutoff : float | None, default=None
180+
The threshold for similarity scores.
181+
perturbation_rate : float | None, default=None
182+
The rate at which to perturb the similarity graph.
183+
delta_threshold : float | None, default=None
184+
The threshold for convergence assessment.
185+
sample_rate : float | None, default=None
186+
The sampling rate for the algorithm.
187+
random_joins : int | None, default=None
188+
The number of random joins to perform.
189+
initial_sampler : str | None, default=None
190+
The initial sampling strategy.
191+
max_iterations : int | None, default=None
192+
The maximum number of iterations to run.
193+
top_k : int | None, default=None
194+
The number of nearest neighbors to find for each node.
195+
random_seed : int | None, default=None
196+
The seed for the random number generator.
197+
concurrency : int | None, default=None
198+
Concurrency configuration.
199+
job_id : str | None, default=None
200+
Job ID for the operation.
201+
log_progress : bool | None, default=None
202+
Whether to log progress.
203+
sudo : bool | None, default=None
204+
Run the algorithm with elevated privileges.
205+
username : str | None, default=None
206+
Username for the operation.
207+
**kwargs : Any
208+
Additional parameters.
209+
210+
Returns
211+
-------
212+
KnnMutateResult
213+
Object containing metadata from the execution.
214+
"""
215+
...

0 commit comments

Comments
 (0)