nystrom-ncut 0.2.0__tar.gz → 0.2.2__tar.gz

{nystrom_ncut-0.2.0/src/nystrom_ncut.egg-info → nystrom_ncut-0.2.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: nystrom_ncut
-Version: 0.2.0
+Version: 0.2.2
 Summary: Normalized Cut and Nyström Approximation
 Author-email: Huzheng Yang <huze.yann@gmail.com>, Wentinn Liao <wentinn.liao@gmail.com>
 Project-URL: Documentation, https://github.com/JophiArcana/Nystrom-NCUT/

{nystrom_ncut-0.2.0 → nystrom_ncut-0.2.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "nystrom_ncut"
-version = "0.2.0"
+version = "0.2.2"
 authors = [
     { name = "Huzheng Yang", email = "huze.yann@gmail.com" },
     { name = "Wentinn Liao", email = "wentinn.liao@gmail.com" },

{nystrom_ncut-0.2.0 → nystrom_ncut-0.2.2}/src/nystrom_ncut/nystrom/nystrom_utils.py

@@ -93,6 +93,7 @@ class OnlineNystrom(TorchTransformerMixin):
 
         self.transform_matrix = (U / L)[:, :self.n_components]                                      # [n x n_components]
         self.eigenvalues_ = L[:self.n_components]                                                   # [n_components]
+        self.is_fitted = True
         return U[:, :self.n_components]                                                             # [n x n_components]
 
     def update(self, features: torch.Tensor) -> torch.Tensor:

{nystrom_ncut-0.2.0 → nystrom_ncut-0.2.2}/src/nystrom_ncut/transformer/axis_align.py

@@ -1,4 +1,3 @@
-import random
 from typing import Literal
 
 import torch
@@ -14,7 +13,7 @@ class AxisAlign(TorchTransformerMixin):
     Args:
         max_iter (int, optional): Maximum number of iterations.
     """
-    SortOptions = Literal["count", "norm"]
+    SortOptions = Literal["count", "norm", "marginal_norm"]
 
     def __init__(
         self,
@@ -33,7 +32,7 @@ class AxisAlign(TorchTransformerMixin):
 
         # Initialize R matrix with the first column from a random row of EigenVectors
         self.R = torch.empty((d, d), device=X.device)
-        self.R[0] = normalized_X[random.randint(0, n - 1)]
+        self.R[0] = normalized_X[torch.randint(0, n, (), device=X.device)]
 
         # Loop to populate R with k orthogonal directions
         c = torch.zeros((n,), device=X.device)
@@ -62,11 +61,16 @@ class AxisAlign(TorchTransformerMixin):
         if self.sort_method == "count":
             sort_metric = torch.bincount(idx, minlength=d)
         elif self.sort_method == "norm":
-            sort_metric = torch.linalg.norm(X @ self.R.mT, dim=0)
+            rotated_X = X @ self.R.mT
+            sort_metric = torch.linalg.norm(rotated_X, dim=0)
+        elif self.sort_method == "marginal_norm":
+            rotated_X = X @ self.R.mT
+            sort_metric = torch.zeros((d,), device=X.device).index_add_(0, idx, rotated_X[range(n), idx] ** 2)
         else:
             raise ValueError(f"Invalid sort method {self.sort_method}.")
 
         self.R = self.R[torch.argsort(sort_metric, dim=0, descending=True)]
+        self.is_fitted = True
         return self
 
     def transform(self, X: torch.Tensor, normalize: bool = True, hard: bool = False) -> torch.Tensor:

{nystrom_ncut-0.2.0 → nystrom_ncut-0.2.2}/src/nystrom_ncut/transformer/transformer_mixin.py

@@ -36,6 +36,8 @@ class TorchTransformerMixin:
     >>> transformer.fit_transform(X)
     array([1, 1, 1])
     """
+    def __init__(self):
+        self.is_fitted: bool = False
 
     @abstractmethod
     def fit(self, X: torch.Tensor, **fit_kwargs: Any) -> "TorchTransformerMixin":

{nystrom_ncut-0.2.0 → nystrom_ncut-0.2.2/src/nystrom_ncut.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: nystrom_ncut
-Version: 0.2.0
+Version: 0.2.2
 Summary: Normalized Cut and Nyström Approximation
 Author-email: Huzheng Yang <huze.yann@gmail.com>, Wentinn Liao <wentinn.liao@gmail.com>
 Project-URL: Documentation, https://github.com/JophiArcana/Nystrom-NCUT/