From fe3b8453b292ea7d68ca16a4e11fabf18557a5d2 Mon Sep 17 00:00:00 2001
From: tomrink <rink@ssec.wisc.edu>
Date: Tue, 23 Apr 2024 13:56:31 -0500
Subject: [PATCH] snapshot...
---
modules/util/hdf5_conversion.py | 10 ++++++++++
1 file changed, 10 insertions(+)
diff --git a/modules/util/hdf5_conversion.py b/modules/util/hdf5_conversion.py
index 3ffef6f9..08dc02e0 100644
--- a/modules/util/hdf5_conversion.py
+++ b/modules/util/hdf5_conversion.py
@@ -1,6 +1,7 @@
import h5py
import numpy as np
import pandas as pd
+import random
def hdf5_to_npz_csv(hdf5_filename, output_file_prefix, chunk_size=10000):
@@ -14,6 +15,15 @@ def hdf5_to_npz_csv(hdf5_filename, output_file_prefix, chunk_size=10000):
chunk_size (int): Size of chunks to process at once (default is 1000).
"""
+ # # New step: keep only 10 mask values == True
+ # mask_indices = list(np.nonzero(mask)[0])
+ # if len(mask_indices) > 10:
+ # selected_indices = random.sample(mask_indices, 10)
+ # new_mask = np.zeros(mask.size).astype(np.bool)
+ # new_mask[selected_indices] = True
+ # mask = new_mask
+ # keep_array.append(data[mask])
+
# Step 1: Open HDF5 file
with h5py.File(hdf5_filename, "r") as file:
fov_mask = np.asarray(file["FOV_mask"])
--
GitLab