More work towards generate_pairs. Now able to pick profiles and compute...

More work towards generate_pairs. Now able to pick profiles and compute euclidian distance between them.

lib/scripts/diffsel_script_utils.py

@@ -127,3 +127,10 @@ def SUBMESSAGE(string):
 
 def WARNING(string):
     myprint(warning_str(string))
+
+def SUCCESS(string):
+    myprint(success_str(string))
+
+def FAILURE(string):
+    myprint(failure_str(string))
+    exit(1)

lib/scripts/generate_pairs.py

@@ -57,6 +57,30 @@ STEP("Reading profiles from file")
 import pandas as pd
 
 MESSAGE("Reading tsv file")
-profiles = pd.read_csv(profiles_file, sep=" ", header=None).transpose()
-print(profiles.shape)
-print(profiles)
+try:
+    profiles = pd.read_csv(profiles_file, sep="\t", header=None).transpose()
+    SUCCESS("Profiles file seems to exist.")
+except:
+    FAILURE("Something went wrong while trying to open profiles file.")
+if profiles.shape[1] != 20:
+    FAILURE("Profiles file does not seem to contain 20 lines.")
+else:
+    SUCCESS("Profiles file contains 20 lines.")
+profiles_sum = profiles.sum(axis=1)
+if max(profiles_sum) < 1.02 and min(profiles_sum) > 0.98:
+    SUCCESS("All profiles sum to 1.")
+else:
+    FAILURE("Some profiles don't sum to 1 (sum is comprised between " + data(min(profiles_sum)) + " and " + data(max(profiles_sum)) + ").")
+MESSAGE("Profiles file contains " + data(profiles.shape[0]) + " profiles.")
+
+#===================================================================================================
+STEP("Picking random pairs")
+from numpy.linalg import norm
+
+def euclidian_distance(p1, p2):
+    return norm(p1 - p2)
+
+def pick(profiles): # returns a profile in the form of a numpy array
+    return profiles.sample().values
+
+print(euclidian_distance(pick(profiles), pick(profiles)))