diff sandbox/embeddings/read.py @ 458:ed6b0b3be8d2

Polished embeddings module
author Joseph Turian <turian@iro.umontreal.ca>
date Tue, 07 Oct 2008 19:13:53 -0400
parents sandbox/embeddings/original.py@131e19dfe793
children f400f62e7f9e
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/sandbox/embeddings/read.py	Tue Oct 07 19:13:53 2008 -0400
@@ -0,0 +1,53 @@
+"""
+Read in the weights file
+"""
+
+import string
+import sys
+
+from files import *
+
+WORDS = None
+WORD_TO_VECTOR = None
+__read = False
+def read_embeddings():
+    global WORDS
+    global WORD_TO_VECTOR
+    global __read 
+    if __read: return
+
+    WORDS = [string.strip(w) for w in open(VOCABFILE).readlines()]
+    assert len(WORDS) == NUMBER_OF_WORDS
+
+    import numpy, math
+    from common.str import percent
+
+    WORD_TO_VECTOR = {}
+
+    sys.stderr.write("Reading %s...\n" % WEIGHTSFILE)
+    f = open(WEIGHTSFILE)
+    f.readline()
+    vals = [float(v) for v in string.split(f.readline())]
+    assert len(vals) == NUMBER_OF_WORDS * DIMENSIONS
+    vals.reverse()
+    for i in range(NUMBER_OF_WORDS):
+        l = vals[DIMENSIONS*i:DIMENSIONS*(i+1)]
+        w = WORDS[i]
+        WORD_TO_VECTOR[w] = l
+    __read = True
+
+import re
+numberre = re.compile("[0-9]")
+def convert_string(s):
+    """
+    Convert a string to a sequence of embeddings.
+    """
+    read_embeddings()
+    e = []
+    for origw in string.split(string.lower(s)):
+        w = numberre.sub("NUMBER", origw)
+        if w in WORD_TO_VECTOR:
+            e.append(WORD_TO_VECTOR[w])
+        else:
+            sys.stderr.write("Word not in vocabulary: %s (original %s)\n" % (w, origw))
+    return e