upload

+-------------- REPORT --------------
+Training Dataset:
+Air: 111
+Gtype: 261
+Gversion: 17
+Med: 119
+Phase: 37
+Strain: 3
+Supp: 294
+Technique: 58
+Temp: 56
+OD: 65
+Anti: 38
+Agit: 4
+Vess: 2
+Substrain: 4
+pH: 26
+
+Mean tags per sentence: 3
+
+
+Testing Dataset:
+Air: 88
+Gtype: 78
+Gversion: 6
+Med: 56
+Phase: 21
+Strain: 0
+Supp: 136
+Technique: 33
+Temp: 15
+OD: 21
+Anti: 13
+Agit: 7
+Vess: 0
+Substrain: 0
+pH: 10
+
+Mean tags per sentence: 4

+# Importacion de librerias
+import pandas as pd
+import numpy as np
+import re
+import argparse
+import os
+
+__author__ = 'kevinml'
+
+# Objective
+# Take CoreNLP processed files and extract two thing.
+# 1.- Number of Each Tag within Training and Testing Datasets
+# 2.- Mean Sentence Tag within Training and Testing Datasets
+
+# Examples
+# python Tags_Training_Testing.py --inputPath /home/kevinml/automatic-extraction-growth-conditions/CRF/data-sets --outputPath /home/kevinml/automatic-extraction-growth-conditions/CRF/analysis_report --TrainingFile training-data-set-70-NER.txt --TestingFile test-data-set-30-NER.txt
+
+####################################################################################
+#                                    MAIN PROGRAM                                  #
+####################################################################################
+
+if __name__ == '__main__':
+    # Definicion de Parametros
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        '--inputPath', help="Ruta donde se encuentran los archivos a procesar. Ej: --inputPath /home/kevinml/automatic-extraction-growth-conditions/data-sets", required=True)
+    parser.add_argument(
+        '--outputPath', help="Ruta donde se depositaran los archivos resultantes. Ej: --outputPath /home/kevinml/automatic-extraction-growth-conditions/data-sets", required=True)
+    parser.add_argument(
+        '--TrainingFile', help="Archivo a procesar. Ej: --TrainingFile test-data-set-30.txt", required=True)
+    parser.add_argument(
+        '--TestingFile', help="Archivo a procesar. Ej: --TestingFile training-data-set-70.txt", required=True)
+    args = parser.parse_args()
+
+    # Se imprimen los parametros ingresados
+    print('\n-------------------------------- PARAMETERS --------------------------------\n')
+    print('Input Path: ' + str(args.inputPath))
+    print('Training File: ' + str(args.TrainingFile))
+    print('Testing File: ' + str(args.TestingFile))
+    print('Output Path: ' + str(args.outputPath))
+    print('\n-------------------------------- PROCESSING --------------------------------\n')
+
+    Air, Gtype, Gversion, Med, Phase, Strain, Supp, Technique, Temp, OD, Anti, Agit, Vess, Substrain, pH = 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    Air_2, Gtype_2, Gversion_2, Med_2, Phase_2, Strain_2, Supp_2, Technique_2, Temp_2, OD_2, Anti_2, Agit_2, Vess_2, Substrain_2, pH_2 = 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0
+    taggs_by_sentence_Testing, taggs_by_sentence_Training = np.array(
+        []), np.array([])
+
+    with open(os.path.join(args.inputPath, args.TrainingFile), mode="r") as Training_file:
+        for line in Training_file:
+            # Descomentar las lineas inferiores para ver cada una de las oraciones en el archivo.
+            # print(line)
+            # print("\n")
+            # Encontramos cuantas veces esta cada etiqueta en el archivo de Training
+            Air += len(re.findall("\|Air", line))
+            Gtype += len(re.findall("\|Gtype", line))
+            Gversion += len(re.findall("\|Gversion", line))
+            Med += len(re.findall("\|Med", line))
+            Phase += len(re.findall("\|Phase", line))
+            Strain += len(re.findall("\|Strain", line))
+            Supp += len(re.findall("\|Supp", line))
+            Technique += len(re.findall("\|Technique", line))
+            Temp += len(re.findall("\|Temp", line))
+            OD += len(re.findall("\|OD", line))
+            Anti += len(re.findall("\|Anti", line))
+            Agit += len(re.findall("\|Agit", line))
+            Vess += len(re.findall("\|Vess", line))
+            Substrain += len(re.findall("\|Substrain", line))
+            pH += len(re.findall("\|pH", line))
+            # Encontramos el promedio de palabras por oracion en el archivo de Training
+            taggs_by_sentence_Training = np.append(taggs_by_sentence_Training, len(re.findall(
+                "\|Air|\|Gtype|\|Gversion|\|Med|\|Phase|\|Strain|\|Supp|\|Technique|\|Temp|\|OD|\|Anti|\|Agit|\|Vess|\|Substrain|\|pH", line)))
+        MeanTags_by_Sent_Training = np.mean(taggs_by_sentence_Training)
+        # Descomentar la linea inferior para ver la lista que contienen el numero de etiquetas en cada una de las oraciones.
+        # print(taggs_by_sentence_Training)
+
+    with open(os.path.join(args.inputPath, args.TestingFile), mode="r") as Testing_file:
+        for line in Testing_file:
+            # Descomentar las lineas inferiores para ver cada una de las oraciones en el archivo.
+            # print(line)
+            # print("\n")
+            # Encontramos cuantas veces esta cada etiqueta en el archivo de Testing
+            Air_2 += len(re.findall("\|Air", line))
+            Gtype_2 += len(re.findall("\|Gtype", line))
+            Gversion_2 += len(re.findall("\|Gversion", line))
+            Med_2 += len(re.findall("\|Med", line))
+            Phase_2 += len(re.findall("\|Phase", line))
+            Strain_2 += len(re.findall("\|Strain", line))
+            Supp_2 += len(re.findall("\|Supp", line))
+            Technique_2 += len(re.findall("\|Technique", line))
+            Temp_2 += len(re.findall("\|Temp", line))
+            OD_2 += len(re.findall("\|OD", line))
+            Anti_2 += len(re.findall("\|Anti", line))
+            Agit_2 += len(re.findall("\|Agit", line))
+            Vess_2 += len(re.findall("\|Vess", line))
+            Substrain_2 += len(re.findall("\|Substrain", line))
+            pH_2 += len(re.findall("\|pH", line))
+            # Encontramos el promedio de palabras por oracion en el archivo de Testing
+            taggs_by_sentence_Testing = np.append(taggs_by_sentence_Testing, len(re.findall(
+                "Air|Gtype|Gversion|Med|Phase|Strain|Supp|Technique|Temp|OD|Anti|Agit|Vess|Substrain|pH", line)))
+        MeanTags_by_Sent_Testing = np.mean(taggs_by_sentence_Testing)
+        # Descomentar la linea inferior para ver la lista que contienen el numero de etiquetas en cada una de las oraciones.
+        # print(taggs_by_sentence_Testing)
+
+    with open(os.path.join(args.outputPath, "Train_Test_tags.txt"), mode="w+") as report:
+        report.write(
+            "-------------- REPORT --------------\nTraining Dataset:\n")
+        report.write("Air: %d\n" % Air)
+        report.write("Gtype: %d\n" % Gtype)
+        report.write("Gversion: %d\n" % Gversion)
+        report.write("Med: %d\n" % Med)
+        report.write("Phase: %d\n" % Phase)
+        report.write("Strain: %d\n" % Strain)
+        report.write("Supp: %d\n" % Supp)
+        report.write("Technique: %d\n" % Technique)
+        report.write("Temp: %d\n" % Temp)
+        report.write("OD: %d\n" % OD)
+        report.write("Anti: %d\n" % Anti)
+        report.write("Agit: %d\n" % Agit)
+        report.write("Vess: %d\n" % Vess)
+        report.write("Substrain: %d\n" % Substrain)
+        report.write("pH: %d\n\n" % pH)
+        report.write("Mean tags per sentence: %d\n\n\n" %
+                     MeanTags_by_Sent_Training)
+
+        report.write("Testing Dataset:\n")
+        report.write("Air: %d\n" % Air_2)
+        report.write("Gtype: %d\n" % Gtype_2)
+        report.write("Gversion: %d\n" % Gversion_2)
+        report.write("Med: %d\n" % Med_2)
+        report.write("Phase: %d\n" % Phase_2)
+        report.write("Strain: %d\n" % Strain_2)
+        report.write("Supp: %d\n" % Supp_2)
+        report.write("Technique: %d\n" % Technique_2)
+        report.write("Temp: %d\n" % Temp_2)
+        report.write("OD: %d\n" % OD_2)
+        report.write("Anti: %d\n" % Anti_2)
+        report.write("Agit: %d\n" % Agit_2)
+        report.write("Vess: %d\n" % Vess_2)
+        report.write("Substrain: %d\n" % Substrain_2)
+        report.write("pH: %d\n\n" % pH_2)
+        report.write("Mean tags per sentence: %d\n" % MeanTags_by_Sent_Testing)
+
+    print("Done.\n")