Adjust PCA to use less memory

catalyst-cooperative · Nov 9, 2023 · e7686c3 · e7686c3
1 parent 1878348
commit e7686c3
Showing 1 changed file with 1 addition and 5 deletions.
diff --git a/src/pudl/analysis/record_linkage/classify_plants_ferc1.py b/src/pudl/analysis/record_linkage/classify_plants_ferc1.py
@@ -7,7 +7,6 @@
 want to do that programmatically, which means using some clustering / categorization
 tools from scikit-learn
 """
-import importlib
 import re
 
 import numpy as np
@@ -214,7 +213,7 @@ def make_ferc1_clf(
             ),
             (
                 "dim_reduction",
-                PCA(n_components=0.99),
+                PCA(copy=False, n_components=1000),
             ),
             (
                 "precompute_dist",
@@ -233,9 +232,6 @@ def make_ferc1_clf(
                 ),
             ),
         ],
-        memory=str(
-            importlib.resources.files("pudl.package_data") / "ferc1_plant_classifier"
-        ),
     )
     return ferc1_pipe