Changeset feed60b in git

Timestamp:

Jul 30, 2019, 12:24:25 AM (4 years ago)

Author:

Murray Heymann <heymann.murray@…>

Branches:

(u'spielwiese', '828514cf6e480e4bafc26df99217bf2a1ed1ef45')

Children:

de88e1d84f18979811df4c3d1dc68f221a3362ef

Parents:

8d8fefe7898bd9e2156488544587049b7107d0c1

Message:

Fix some minor errors

Location:

machine_learning

Files:

• common/keyword_vector.py (modified) (3 diffs)
• common/lookuptable.py (modified) (1 diff)
• model/predictor.py (modified) (8 diffs)

machine_learning/common/keyword_vector.py

@@ -3 +3 @@
 """Some vector logic"""
 
-import numpy as np
 import os
 import re
 import sys
-import traceback
+import numpy as np
 
 from common.constants import KEYWORDS_FILE
@@ -19 +18 @@
     if not os.path.isfile(filename):
         print("Please provide a valid input file as argument")
-        return []
+        return np.array([])
 
     dictionary = []
@@ -148 +147 @@
 
     print("Attempt to normalise the zero vector")
-    print(normalise_vector(np.array([0,0,0,0,0])))
+    print(normalise_vector(np.array([0, 0, 0, 0, 0])))
     print()
 
     print("Attempt to normalise list")
-    print(normalise_vector([3,4,0,0,0]))
+    print(normalise_vector([3, 4, 0, 0, 0]))
     print()
 

machine_learning/common/lookuptable.py

@@ -120 +120 @@
     test_vec = count_occurances(os.path.join(HELP_FILE_PATH, "html",
                                              files[1]), dictionary)
-    print((test_vec==vectors[1]).all())
+    print((test_vec == vectors[1]).all())
 
 

machine_learning/model/predictor.py

@@ -3 +3 @@
 """
 
-import cProfile
+# import cProfile
 import os
 import sys
@@ -56 +56 @@
                 # dist = vector_distance(x, vec)
                 # Dot product is much faster
-                dist = -np.dot(x, vec)
+                dist = -1 * np.dot(x, vec)
                 if dist < min_val:
                     min_val = dist
@@ -68 +68 @@
 
 
-def main():
+def basic_vector_tests():
     """
-    Run some basic tests
+    Some basic sanity tests
     """
-    print("Running some tests")
     predictor = HelpPagePredictor()
-    vector1 = normalise_vector([1, 4, 10])
-    vector2 = normalise_vector([2, 3, 1])
-    vector3 = normalise_vector([3, 9, 3])
+    vector1 = normalise_vector(np.array([1, 4, 10]))
+    vector2 = normalise_vector(np.array([2, 3, 1]))
+    vector3 = normalise_vector(np.array([3, 9, 3]))
 
     vectors = np.array([vector1, vector2, vector3])
@@ -84 +83 @@
     print()
 
-    testvec = normalise_vector([1, 1, 1])
+    testvec = normalise_vector(np.array([1, 1, 1]))
     print("test vector:")
     print(testvec)
@@ -105 +104 @@
     print()
 
+
+def main():
+    """
+    Run some basic tests
+    """
+    print("Running some tests")
+
+    basic_vector_tests()
+
     dictionary = read_dictionary(KEYWORDS_FILE)
 
@@ -113 +121 @@
 
     test_vec = count_occurances("extract.lib", dictionary)
+    predictor = HelpPagePredictor()
     predictor.fit(vectors, file_list)
 
@@ -124 +133 @@
     print("prediction for zero vector")
     zerovec = np.zeros(len(dictionary) - 2)
-    print(len(zerovec))
     start = time.time()
     prediction = predictor.predict(np.array([zerovec]))
@@ -138 +146 @@
             if not os.path.isfile(sys.argv[i]):
                 continue
-            print ("predicting for file", sys.argv[i])
+            print("predicting for file", sys.argv[i])
             test_vec = count_occurances(sys.argv[i], dictionary)
             start = time.time()