Feature extraction and vectorizer three sentences

@@ -19,16 +19,17 @@ __author__ = 'CMendezC'
 # 3) --vectorizer Vectorizer: b=binary, f=frequency, t=tf-idf.
 
 # Ouput:
-# 1) Files with vectors.
+# 1) Report with dictionary, vectors, cosine similarity matrix.
 
 # Execution:
 # python extraccion-caracteristicas-vectorizacion.py
 # --inputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/data-set-three-sentences
 # --outputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/reports-three-sentences
 # --vectorizer b
+# --feature word
 
 # source activate python3
-# python extraccion-caracteristicas-vectorizacion.py --inputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/data-set-three-sentences --outputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/reports-three-sentences --vectorizer b
+# python extraccion-caracteristicas-vectorizacion.py --inputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/data-set-three-sentences --outputPath /home/compu2/bionlp/lcg-bioinfoI-bionlp/representaciones-vectoriales/reports-three-sentences --vectorizer b --feature word
 
 ###########################################################
 #                       MAIN PROGRAM                      #
@@ -44,6 +45,9 @@ if __name__ == "__main__":
     parser.add_argument("--vectorizer", dest="vectorizer", required=True,
                       help="Vectorizer: b=binary, f=frequency, t=tf-idf", metavar="CHAR",
                       choices=('b', 'f', 't'), default='b')
+    parser.add_argument("--feature", dest="feature", required=True,
+                      help="Feature: word, lemma, pos", metavar="TEXT",
+                      choices=('word', 'lemma', 'pos'), default='b')
 
     args = parser.parse_args()
 
@@ -61,10 +65,11 @@ if __name__ == "__main__":
     # Read documents from input path
     for path, dirs, files in os.walk(args.inputPath):
         for file in files:
-            with open(os.path.join(args.inputPath, file), mode="r", encoding="utf-8") as iFile:
-                print("...{}".format(file))
-                # Add file to document list
-                documents.append(iFile.read())
+            if file.endswith(args.feature):
+                with open(os.path.join(args.inputPath, file), mode="r", encoding="utf-8") as iFile:
+                    print("...{}".format(file))
+                    # Add file to document list
+                    documents.append(iFile.read())
     print("  Documents: {}".format(len(documents)))
 
     # Create vectorizer
@@ -85,7 +90,7 @@ if __name__ == "__main__":
     similarityMatrix = cosine_similarity(matrix)
     print("   Cosine similarity matrix shape: {}".format(similarityMatrix.shape))
 
-    with open(os.path.join(args.outputPath, "report-vectorizer.{}.txt".format(args.vectorizer)), encoding="utf-8", mode="w") as oFile:
+    with open(os.path.join(args.outputPath, "report-vectorizer.{}.{}.txt".format(args.feature, args.vectorizer)), encoding="utf-8", mode="w") as oFile:
         oFile.write("Vectorizer: {}\n".format(args.vectorizer))
         oFile.write(str(vectorizer.get_feature_names()))
         oFile.write("\n")