From 86804955ed180c54a83660c3cace37f27fe14027 Mon Sep 17 00:00:00 2001
From: tomrink <rink@ssec.wisc.edu>
Date: Thu, 20 Oct 2022 15:48:44 -0500
Subject: [PATCH] snapshot...
---
modules/deeplearning/srcnn_l1b_l2.py | 13 ++++++++-----
1 file changed, 8 insertions(+), 5 deletions(-)
diff --git a/modules/deeplearning/srcnn_l1b_l2.py b/modules/deeplearning/srcnn_l1b_l2.py
index 33a18220..94aa3b36 100644
--- a/modules/deeplearning/srcnn_l1b_l2.py
+++ b/modules/deeplearning/srcnn_l1b_l2.py
@@ -49,8 +49,8 @@ f.close()
mean_std_dct.update(mean_std_dct_l1b)
mean_std_dct.update(mean_std_dct_l2)
-label_param = 'cloud_fraction'
-# label_param = 'cld_opd_dcomp'
+# label_param = 'cloud_fraction'
+label_param = 'cld_opd_dcomp'
# label_param = 'cloud_probability'
params = ['temp_11_0um_nom', 'temp_12_0um_nom', 'refl_0_65um_nom', label_param]
@@ -251,7 +251,8 @@ class SRCNN:
# label = input_data[:, label_idx, 3:131:2, 3:131:2]
label = input_data[:, label_idx, 3:131, 3:131]
if label_param != 'cloud_fraction':
- label = normalize(label, label_param, mean_std_dct)
+ #label = normalize(label, label_param, mean_std_dct)
+ label = np.where(np.isnan(label), 0, label)
else:
label = np.where(np.isnan(label), 0, label)
label = np.expand_dims(label, axis=3)
@@ -399,9 +400,9 @@ class SRCNN:
conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_3', scale=scale)
- conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_4', scale=scale)
+ # conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_4', scale=scale)
- conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_5', scale=scale)
+ # conv_b = build_residual_conv2d_block(conv_b, num_filters, 'Residual_Block_5', scale=scale)
conv_b = tf.keras.layers.Conv2D(num_filters, kernel_size=3, strides=1, kernel_initializer='he_uniform', padding=padding)(conv_b)
@@ -646,6 +647,8 @@ class SRCNN:
def run(self, directory, ckpt_dir=None, num_data_samples=50000):
train_data_files = glob.glob(directory+'data_train_*.npy')
valid_data_files = glob.glob(directory+'data_valid_*.npy')
+ train_data_files = train_data_files[::2]
+ valid_data_files = valid_data_files[::2]
self.setup_pipeline(train_data_files, valid_data_files, num_data_samples)
self.build_model()
--
GitLab