Changeset 746d3d in git

Timestamp:

Jul 26, 2019, 3:31:01 PM (4 years ago)

Author:

Murray Heymann <heymann.murray@…>

Branches:

(u'spielwiese', '828514cf6e480e4bafc26df99217bf2a1ed1ef45')

Children:

8720894b787185ed212506a9a4492ddf6e5f014c

Parents:

dd499fe3f7367342e4f0cf3495cdab72a21c2280

git-author:

Murray Heymann <heymann.murray@gmail.com>2019-07-26 15:31:01+02:00

git-committer:

Murray Heymann <heymann.murray@gmail.com>2019-07-26 15:31:07+02:00

Message:

Write lookup table of vectors and  their helpfiles

Files:

• .gitignore (modified) (1 diff)
• machine_learning/common/constants.py (added)
• machine_learning/common/keyword_vector.py (modified) (3 diffs)
• machine_learning/common/lookuptable.py (added)
• machine_learning/common/predictor.py (modified) (4 diffs)
• machine_learning/extract.lib (modified) (1 diff)

.gitignore

@@ -118 +118 @@
 /resources/singular_resourcesconfig.h
 __pycache__
+keywords.txt
+helpfiles

machine_learning/common/keyword_vector.py

@@ -126 +126 @@
 
 
-if __name__ == '__main__':
+def main():
+    """
+    Run some basic tests
+    """
     if len(sys.argv) != 2:
         print("Usage: ")
@@ -132 +135 @@
         sys.exit(1)
 
-    dic = read_dictionary(sys.argv[1]) # pylint: disable=invalid-name
+    dic = read_dictionary(sys.argv[1])
 
-    testvector = {"hello":3, "bye":4} # pylint: disable=invalid-name
+    testvector = {"hello":3, "bye":4}
     normalise_vector(testvector)
     print("normalised vector: " + str(testvector))
 
-    vector1 = {"hello":3, "bye":4} # pylint: disable=invalid-name
-    vector2 = {"hello":4, "bye":3} # pylint: disable=invalid-name
+    vector1 = {"hello":3, "bye":4}
+    vector2 = {"hello":4, "bye":3}
     print("distance same vector: " + str(vector_distance(vector1, vector1)))
     print("distance different vector: " + str(vector_distance(vector1, vector2)))
@@ -146 +149 @@
 
     print(count_occurances("../Singular/table.h", dic))
+
+if __name__ == '__main__':
+    main()

machine_learning/common/predictor.py

@@ -22 +22 @@
         self.files = None
 
+
     def fit(self, X, y): # pylint: disable=invalid-name
         """
@@ -31 +32 @@
         self.files = y
         return self
+
 
     def predict(self, X): # pylint: disable=invalid-name
@@ -56 +58 @@
 
 
-if __name__ == '__main__':
+def main():
+    """
+    Run some basic tests
+    """
     print("Running some tests")
-    predictor = HelpPagePredictor() # pylint: disable=invalid-name
-    vector1 = {"hello":1, "bye":4, "pizza": 10} # pylint: disable=invalid-name
-    vector2 = {"hello":2, "bye":3, "pizza": 1} # pylint: disable=invalid-name
-    vector3 = {"hello":3, "bye":9, "pizza": 3} # pylint: disable=invalid-name
+    predictor = HelpPagePredictor()
+    vector1 = {"hello":1, "bye":4, "pizza": 10}
+    vector2 = {"hello":2, "bye":3, "pizza": 1}
+    vector3 = {"hello":3, "bye":9, "pizza": 3}
 
-    vectors = np.array([vector1, vector2, vector3]) # pylint: disable=invalid-name
-    files = np.array(["file1", "file2", "file3"]) # pylint: disable=invalid-name
+    vectors = np.array([vector1, vector2, vector3])
+    files = np.array(["file1", "file2", "file3"])
     print(vectors)
     print(files)
 
-    testvec = {"hello":1, "bye":1, "pizza": 1} # pylint: disable=invalid-name
+    testvec = {"hello":1, "bye":1, "pizza": 1}
 
     print("distance to 1")
@@ -81 +86 @@
 
     predictor.fit(vectors, files)
-    prediction = predictor.predict(np.array([testvec])) # pylint: disable=invalid-name
+    prediction = predictor.predict(np.array([testvec]))
     print(prediction)
+
+if __name__ == '__main__':
+    main()

machine_learning/extract.lib

@@ -6 +6 @@
 
 // create file, overwrite if exists
-write(":w test.txt", l[1]);
+write(":w keywords.txt", l[1]);
 
 // write entries from "names" procedure
 for (i = 1; i < size(l); i++) {
-        write(":a test.txt", l[i+1]);
+        write(":a keywords.txt", l[i+1]);
 }
 
 // write entries from "reservedNameList" procedure
 for (i = 1; i < size(k) + 1; i++) {
-        write(":a test.txt", l[i]);
+        write(":a keywords.txt", l[i]);
 }
+
+exit;