snapshot...

modules/deeplearning/srcnn_l1b_l2.py

@@ -7,6 +7,7 @@ import os, datetime
 import numpy as np
 import pickle
 import h5py
+from scipy.ndimage import gaussian_filter
 
 # L1B M/I-bands: /apollo/cloud/scratch/cwhite/VIIRS_HRES/2019/2019_01_01/
 # CLAVRx: /apollo/cloud/scratch/Satellite_Output/VIIRS_HRES/2019/2019_01_01/
@@ -754,6 +755,7 @@ def run_evaluate_static(in_file, out_file, ckpt_dir):
     # grd_b = normalize(grd_b, 'refl_0_65um_nom', mean_std_dct)
 
     grd_c = get_grid_values_all(h5f, label_param)
+    # grd_c = gaussian_filter(grd_c, sigma=1.0)
     grd_c = grd_c[y_0:y_0+sub_y, x_0:x_0+sub_x]
     hr_grd_c = grd_c.copy()
     hr_grd_c = hr_grd_c[y_128, x_128]
@@ -777,39 +779,6 @@ def run_evaluate_static(in_file, out_file, ckpt_dir):
         return out_sr, None, None
 
 
-def analyze(fpath='/Users/tomrink/clavrx_snpp_viirs.A2019080.0100.001.2019080064252.uwssec_B00038315.level2.h5', param='cloud_probability'):
-    h5f = h5py.File(fpath, 'r')
-    grd = get_grid_values_all(h5f, param)
-    grd = np.where(np.isnan(grd), 0, grd)
-    bt = get_grid_values_all(h5f, 'temp_11_0um_nom')
-    refl = get_grid_values_all(h5f, 'refl_0_65um_nom')
-    grd = grd[2432:4032, 2432:4032]
-    bt = bt[2432:4032, 2432:4032]
-    refl = refl[2432:4032, 2432:4032]
-    print(grd.shape)
-
-    grd_lr = grd[::2, ::2]
-    print(grd_lr.shape)
-    leny, lenx = grd_lr.shape
-    rnd = np.random.normal(loc=0, scale=0.001, size=grd_lr.size)
-    grd_lr = grd_lr + rnd.reshape(grd_lr.shape)
-    if param == 'cloud_probability':
-        grd_lr = np.where(grd_lr < 0, 0, grd_lr)
-        grd_lr = np.where(grd_lr > 1, 1, grd_lr)
-
-    x = np.arange(lenx)
-    y = np.arange(leny)
-    x_up = np.arange(0, lenx, 0.5)
-    y_up = np.arange(0, leny, 0.5)
-
-    grd_hr = resample_2d_linear_one(x, y, grd_lr, x_up, y_up)
-    print(grd_hr.shape)
-
-    h5f.close()
-
-    return grd, grd_lr, grd_hr, bt, refl
-
-
 if __name__ == "__main__":
     nn = SRCNN()
     nn.run('matchup_filename')