Final version binClass for papers

@@ -3,7 +3,8 @@ from sklearn.cross_validation import train_test_split as splitt
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.feature_extraction.text import TfidfVectorizer
 from sklearn.model_selection import RandomizedSearchCV
 from sklearn.model_selection import RandomizedSearchCV
 from sklearn.model_selection import GridSearchCV
 from sklearn.model_selection import GridSearchCV
-from sklearn import metrics
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import recall_score, precision_score, f1_score, classification_report
 from sklearn.svm import SVC
 from sklearn.svm import SVC
 import numpy as np
 import numpy as np
 import argparse
 import argparse
@@ -16,18 +17,23 @@ from scipy.stats import expon
 from sklearn.preprocessing import label_binarize
 from sklearn.preprocessing import label_binarize
 from sklearn.datasets import load_files
 from sklearn.datasets import load_files
 
 
+# CMC: Run example
+# python3.4 filter_papers.py --traind /home/cmendezc/gitlab_repositories/useless/data/TEXT_FILES
 
 
 parser = argparse.ArgumentParser(
 parser = argparse.ArgumentParser(
     description="This script separates biomedical papers that"
     description="This script separates biomedical papers that"
             "report data from biomedical experiments from those that do not.")
             "report data from biomedical experiments from those that do not.")
-parser.add_argument("--input", help="Input file containing the  to"
+parser.add_argument("--input", help="Input directory containing the papers to"
                                 "be predited.")
                                 "be predited.")
 parser.add_argument("--traind", help="Input directory containing the papers of"
 parser.add_argument("--traind", help="Input directory containing the papers of"
                                 "two classes to be learned.")
                                 "two classes to be learned.")
 parser.add_argument("--out", help="Path to the output directory "
 parser.add_argument("--out", help="Path to the output directory "
                      "(default='./filter_output')", default="filter_output")
                      "(default='./filter_output')", default="filter_output")
 parser.add_argument("--svcmodel", help="Path to custom pretrained svc model"
 parser.add_argument("--svcmodel", help="Path to custom pretrained svc model"
-        "(default='./model/svm_model.paper.pkl')", default="model/svm_model.paper.pkl")
+        "(default='./model_binClass/svm_model.paper.pkl')", default="model_binClass/svm_model.paper.pkl")
+parser.add_argument("--split", default=False,
+                  action="store_true", dest="split",
+                  help="Automatic split training/test of input data ")
 
 
 args = parser.parse_args()
 args = parser.parse_args()
 labels = {0: 'useless', 1: 'useful'}
 labels = {0: 'useless', 1: 'useful'}
@@ -56,10 +62,19 @@ if args.traind and not args.input:
 
 
     model_params = {k: list(set(model_params[k])) for k in model_params}
     model_params = {k: list(set(model_params[k])) for k in model_params}
 
 
-    tfidf_model = vectorizer.fit(data.data)
-    X = vectorizer.transform(data.data)
-    #y = [x['topic'] for x in abstracs]
-    y = data.target    
+    # CMC: separate in training - validation datasets
+    if args.split:
+        X_train, X_test, y_train, y_test = train_test_split(data.data, labels, test_size = 0.25, random_state = 42)
+        tfidf_model = vectorizer.fit(X_train)
+        X = vectorizer.transform(X_train)
+        y = y_train
+    else:
+        #y = [x['topic'] for x in abstracs]
+        # Original Nacho:
+        tfidf_model = vectorizer.fit(data.data)
+        X = vectorizer.transform(data.data)
+        y = data.target
+
 
 
     #X_train, X_test, y_train, y_test = splitt(X, y, test_size=0.3, random_state=42)
     #X_train, X_test, y_train, y_test = splitt(X, y, test_size=0.3, random_state=42)
 
 
@@ -81,15 +96,22 @@ if args.traind and not args.input:
     #print(metrics.f1_score(clf.predict(X_test), y_test))
     #print(metrics.f1_score(clf.predict(X_test), y_test))
 
 
     #joblib.dump(clf, 'model/svm_model.pkl')
     #joblib.dump(clf, 'model/svm_model.pkl')
-    joblib.dump(clf.best_estimator_, 'model/svm_model.paper.pkl')
-    joblib.dump(tfidf_model, 'model/tfidf_model.paper.pkl')
+    joblib.dump(clf.best_estimator_, 'model_binClass/svm_model.paper.pkl')
+    joblib.dump(tfidf_model, 'model_binClass/tfidf_model.paper.pkl')
 
 
+    if args.split:
+        X = vectorizer.transform(X_test)
+        y_pred = clf.predict(X)
+        print(precision_score(y_test, y_pred))
+        print(recall_score(y_test, y_pred))
+        print(f1_score(y_test, y_pred))
+        print(classification_report(y_test, y_pred))
 else:
 else:
     from pdb import set_trace as st
     from pdb import set_trace as st
     data=load_files(container_path=args.input, encoding=None,
     data=load_files(container_path=args.input, encoding=None,
                                                     decode_error='replace')
                                                     decode_error='replace')
     clf = joblib.load(args.svcmodel)
     clf = joblib.load(args.svcmodel)
-    vectorizer = joblib.load('model/tfidf_model.paper.pkl')
+    vectorizer = joblib.load('model_binClass/tfidf_model.paper.pkl')
     X = vectorizer.transform(data.data)
     X = vectorizer.transform(data.data)
 
 
     classes = clf.predict(X)
     classes = clf.predict(X)
@@ -97,10 +119,10 @@ else:
     if not os.path.exists(args.out):
     if not os.path.exists(args.out):
         os.makedirs(args.out)
         os.makedirs(args.out)
     # Writing predictions to output files
     # Writing predictions to output files
-    with open(args.out + "/" + labels[0] + ".out", 'w') as f0, \
-                    open(args.out + "/" + labels[1] + ".out", 'w') as f1:
-        for c, a in zip(classes, papers):
+    with open(args.out + "/" + labels[0] + "-binClass-paper.out", 'w') as f0, \
+                    open(args.out + "/" + labels[1] + "-binClass-paper.out", 'w') as f1:
+        for c, a in zip(classes, data):
             if c == 0:
             if c == 0:
-                f0.write("%d\t%s\n" % (a['title'], a['body']))
+                f0.write("%d\n" % (a['title']))
             elif c == 1:
             elif c == 1:
-                f1.write("%d\t%s\n" % (a['title'], a['body']))
+                f1.write("%d\n" % (a['title']))