more tests for distance. All tests pass now

.gitignore

+*__pycache__
\ No newline at end of file

distance_batsim_output.py

@@ -66,10 +66,14 @@ def lateness_distance(s1, s2):
     return np.sum([y-x for x, y in zip(s1, s2)])
 
 def normalized_euclidian_distance(s1, s2):
-    """Return the euclidien distance normalized by the l2 norm of the vectors"""
+    """Return the euclidien distance normalized by the l2 norm of the vectors, 
+    or None if one of the vectors is the null vector (undefined)"""
 
+    n1, n2 = l2_norm(s1), l2_norm(s2)
+    if n1==0 or n2==0:
+        return None
     eucl_dist = euclidean_distance(s1, s2)
-    return eucl_dist**2 / (l2_norm(s1) * l2_norm(s2))
+    return eucl_dist**2 / (n1 * n2)
 
 def l2_norm(s):
     """Return the l2 norm of the series s"""

test/input/3jobs.csv

+job_id,workload_name,profile,submission_time,requested_number_of_resources,requested_time,success,final_state,starting_time,execution_time,finish_time,waiting_time,turnaround_time,stretch,allocated_resources,consumed_energy,metadata
+1216,user11,362,30,1,86400.000000,1,COMPLETED_SUCCESSFULLY,30,362.000000,80,0.000000,362.000000,1.000000,2,62671.250000,
+247,user5,57102,0,8,432000.000000,1,COMPLETED_SUCCESSFULLY,0,57102.000000,0,0.000000,57102.000000,1.000000,0,12391134.000000,
+1242,user11,9620,40,1,86400.000000,1,COMPLETED_SUCCESSFULLY,40,9620.000000,60,0.000000,9620.000000,1.000000,2,1665462.500000,
\ No newline at end of file

test/test_distance.py

@@ -32,6 +32,7 @@ def test_normalized_euclidean_distance():
 
 
 ####### Integration tests #######
+three_jobs = "test/input/3jobs.csv"
 three_jobs_w_session = "test/input/3jobs_w_sessions.csv"
 three_jobs_zero = "test/input/3jobs_zeros.csv"
 three_jobs_one_unsuccessful = "test/input/3jobs_1unsuccessful.csv"
@@ -43,38 +44,43 @@ def test_cleaning():
     
     # Clean unsuccessful jobs:
     with pytest.warns(UserWarning):
-        distances(three_jobs_w_session, three_jobs_one_unsuccessful)
+        distances(three_jobs, three_jobs_one_unsuccessful)
 
     # Complain if no matching job_ids:
     with pytest.raises(KeyError):
-        distances(three_jobs_w_session, mc_10days_a60)
+        distances(three_jobs, mc_10days_a60)
 
 
 def test_some_distances():
     fin, sub, start = ["finish_time"], ["submission_time"], ["starting_time"]
 
     # d(u,u) == 0 for all distances
-    assert distances(three_jobs_w_session, three_jobs_w_session, 
-        field=sub) == 0 
-    assert distances(three_jobs_w_session, three_jobs_w_session, 
-        field=fin) == 0 
-    assert distances(three_jobs_w_session, three_jobs_w_session, 
-        euclidean=False, norm_eucl=True, field=fin) == 0 
-    assert distances(three_jobs_w_session, three_jobs_w_session, 
-        euclidean=False, lateness=True, field=fin) == 0 
+    assert distances(three_jobs, three_jobs, field=sub) == 0 
+    assert distances(three_jobs, three_jobs, field=fin) == 0 
+    assert distances(three_jobs, three_jobs, 
+        euclidean=False, norm_eucl=True, field=fin)     == 0 
+    assert distances(three_jobs, three_jobs, 
+        euclidean=False, lateness=True, field=fin)      == 0 
 
     # Eucl distance
-    assert distances(three_jobs_w_session, three_jobs_zero,
-        field=sub) == 50
-    assert distances(three_jobs_zero, three_jobs_w_session,
-        field=start) == 50
-    assert distances(three_jobs_w_session, three_jobs_zero,
-        field=fin) == 100
+    assert distances(three_jobs, three_jobs_zero, field=sub)    == 50
+    assert distances(three_jobs_zero, three_jobs, field=start)  == 50
+    assert distances(three_jobs, three_jobs_zero, field=fin)    == 100
+
+    assert distances(three_jobs, three_jobs_w_session, field=sub)   == 0
+    assert distances(three_jobs, three_jobs_w_session, field=start) == 0
+    assert distances(three_jobs, three_jobs_w_session, field=fin)   == 20 * np.sqrt(2)
 
     # Normalized eucl distance
-    assert distances(three_jobs_w_session, three_jobs_zero, 
-        euclidean=False, norm_eucl=True, field=sub) ==          1
-    assert distances(three_jobs_zero, three_jobs_w_session,
-        euclidean=False, norm_eucl=True, field=start) ==        1
-    assert distances(three_jobs_w_session, three_jobs_zero,
-        euclidean=False, norm_eucl=True, field=fin) ==          1
\ No newline at end of file
+    assert distances(three_jobs, three_jobs_zero, 
+        euclidean=False, norm_eucl=True, field=sub)           == None
+    assert distances(three_jobs, three_jobs_w_session,
+        euclidean=False, norm_eucl=True, field=sub)           == 0
+    
+    norm_dis_A_B = distances(three_jobs, three_jobs_w_session,
+                        euclidean=False, norm_eucl=True, field=fin)
+    norm_dis_B_A = distances(three_jobs, three_jobs_w_session,
+                        euclidean=False, norm_eucl=True, field=fin)
+    expected = 800 / (100*100)
+    assert norm_dis_A_B == norm_dis_B_A
+    assert norm_dis_B_A - expected  < 1e-8