snapshot...

modules/deeplearning/cnn_cld_frac.py

@@ -1032,19 +1032,21 @@ def analyze(file='/Users/tomrink/cld_opd_out.npy'):
 
 
 def analyze2(nda_m, nda_i):
+    n_imgs = nda_m.shape[0]
     nda_m = np.where(nda_m < 0.5, 0, 1)
     nda_i = np.where(nda_i < 0.5, 0, 1)
 
-    cf_m = np.zeros((64, 64))
-    cf_i = np.zeros((64, 64))
+    cf_m = np.zeros((n_imgs, 64, 64))
+    cf_i = np.zeros((n_imgs, 64, 64))
 
-    for j in range(1, 65):
-        for i in range(1, 65):
-            sub_3x3 = nda_m[j-1:j+2, i-1:i+2]
-            cf_m[j-1, i-1] = np.sum(sub_3x3)
+    for k in range(n_imgs):
+        for j in range(1, 65):
+            for i in range(1, 65):
+                sub_3x3 = nda_m[k, j-1:j+2, i-1:i+2]
+                cf_m[k, j-1, i-1] = np.sum(sub_3x3)
 
-            sub_4x4 = nda_i[j*2-1:j*2+3, i*2-1:i*2+3]
-            cf_i[j-1, i-1] = np.sum(sub_4x4)
+                sub_4x4 = nda_i[k, j*2-1:j*2+3, i*2-1:i*2+3]
+                cf_i[k, j-1, i-1] = np.sum(sub_4x4)
 
     # cat_0 = cf_m == 0
     # cat_1 = (cf_m >= 0.1) & (cf_m < 0.13)
@@ -1068,21 +1070,22 @@ def analyze2(nda_m, nda_i):
     # cf_m[cat_8] = 8
     # cf_m[cat_9] = 9
 
-    cat_0 = (cf_m == 0)
-    cat_1 = (cf_m > 0) & (cf_m < 9)
-    cat_2 = cf_m == 9
+    for k in range(n_imgs):
+        cat_0 = (cf_m[k, ] == 0)
+        cat_1 = (cf_m[k, ] > 0) & (cf_m[k, ] < 9)
+        cat_2 = cf_m[k, ] == 9
 
-    cf_m[cat_0] = 0
-    cf_m[cat_1] = 1
-    cf_m[cat_2] = 2
+        cf_m[k, cat_0] = 0
+        cf_m[k, cat_1] = 1
+        cf_m[k, cat_2] = 2
 
-    cat_0 = (cf_i == 0)
-    cat_1 = (cf_i > 0) & (cf_i < 16)
-    cat_2 = cf_i == 16
+        cat_0 = (cf_i[k, ] == 0)
+        cat_1 = (cf_i[k, ] > 0) & (cf_i[k, ] < 16)
+        cat_2 = cf_i[k, ] == 16
 
-    cf_i[cat_0] = 0
-    cf_i[cat_1] = 1
-    cf_i[cat_2] = 2
+        cf_i[k, cat_0] = 0
+        cf_i[k, cat_1] = 1
+        cf_i[k, cat_2] = 2
 
     return cf_m, cf_i