From 6ea5cb6f886fab2ed80bc26422dd152282626160 Mon Sep 17 00:00:00 2001
From: Feng Tian <134106614+ftian9@users.noreply.github.com>
Date: Wed, 24 Jun 2026 09:21:58 -0700
Subject: [PATCH 1/3] skip downloading the moments file if it already exists

---
 molskill/data/standardization.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/molskill/data/standardization.py b/molskill/data/standardization.py
index 2f2ab19..2075425 100644
--- a/molskill/data/standardization.py
+++ b/molskill/data/standardization.py
@@ -29,7 +29,7 @@ def get_population_moments(
         Dict[str, np.ndarray]: population {mean: np.ndarray, std: np.ndarray}
     """
 
-    if moment_csv is None:
+    if moment_csv is None and not os.path.exists(MOMENT_CSV):
         LOGGER.info("Standardization moments not found. Downloading from remote...")
         moment_csv = MOMENT_CSV
         os.makedirs(ASSET_PATH, exist_ok=True)

From a53bb105b4ebfed2afaebe5067e1bd341e16cbc0 Mon Sep 17 00:00:00 2001
From: Feng Tian <134106614+ftian9@users.noreply.github.com>
Date: Wed, 24 Jun 2026 09:32:08 -0700
Subject: [PATCH 2/3] check moments file size

---
 molskill/data/standardization.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/molskill/data/standardization.py b/molskill/data/standardization.py
index 2075425..266e76e 100644
--- a/molskill/data/standardization.py
+++ b/molskill/data/standardization.py
@@ -29,7 +29,7 @@ def get_population_moments(
         Dict[str, np.ndarray]: population {mean: np.ndarray, std: np.ndarray}
     """
 
-    if moment_csv is None and not os.path.exists(MOMENT_CSV):
+    if moment_csv is None and (not os.path.exists(MOMENT_CSV)) or (os.path.getsize(MOMENT_CSV) == 0):
         LOGGER.info("Standardization moments not found. Downloading from remote...")
         moment_csv = MOMENT_CSV
         os.makedirs(ASSET_PATH, exist_ok=True)

From 9462620c944af9cf18484e4e37c14e7af8dc9242 Mon Sep 17 00:00:00 2001
From: Feng Tian <134106614+ftian9@users.noreply.github.com>
Date: Wed, 24 Jun 2026 09:46:15 -0700
Subject: [PATCH 3/3] update variable moment_csv if it is None

---
 molskill/data/standardization.py | 9 +++++----
 1 file changed, 5 insertions(+), 4 deletions(-)

diff --git a/molskill/data/standardization.py b/molskill/data/standardization.py
index 266e76e..fd85542 100644
--- a/molskill/data/standardization.py
+++ b/molskill/data/standardization.py
@@ -29,11 +29,12 @@ def get_population_moments(
         Dict[str, np.ndarray]: population {mean: np.ndarray, std: np.ndarray}
     """
 
-    if moment_csv is None and (not os.path.exists(MOMENT_CSV)) or (os.path.getsize(MOMENT_CSV) == 0):
-        LOGGER.info("Standardization moments not found. Downloading from remote...")
+    if moment_csv is None:
         moment_csv = MOMENT_CSV
-        os.makedirs(ASSET_PATH, exist_ok=True)
-        download(DEFAULT_MOMENTS_REMOTE, moment_csv)
+        if (not os.path.exists(moment_csv)) or (os.path.getsize(moment_csv) == 0):
+            LOGGER.info("Standardization moments not found. Downloading from remote...")
+            os.makedirs(ASSET_PATH, exist_ok=True)
+            download(DEFAULT_MOMENTS_REMOTE, moment_csv)
 
     population_df = pd.read_csv(moment_csv, index_col="descriptor")
     assert set(population_df.keys()) == {