Added missing files

src/backend/classifier.py

@@ -0,0 +1,37 @@
+from tables_manager import Tables
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import f1_score
+import logging
+import pandas as pd
+
+class Classifier:
+        def __init__(self):
+                self.tables_manager = Tables()
+                self.data = self.tables_manager.get_all_matches().dropna()
+                logging.info(self.data.head())
+                self.x = self.data[['R1','R2','R3','R4','R5','D1','D2','D3','D4','D5']].values
+                self.y = self.data[['radiant_win']].values
+                self.test_size=0.3
+                self.x_train, self.x_test, self.y_train, self.y_test = train_test_split(self.x, self.y, test_size=self.test_size)
+                self.classifier = None #To be filled by child class.
+
+        def train(self):
+                self.classifier.fit(self.x_train, self.y_train.ravel())
+
+        def results(self):
+                self.result = self.classifier.predict(self.x_test)
+                self.confusion_matrix = pd.crosstab(self.y_test.ravel(),self.result, rownames=['Real'], colnames=['Predito'], margins=True)
+                self.fscore = f1_score(self.y_test.ravel(),self.result,average='macro')
+                self.overview = {'F-score': self.fscore, 'Confusion-matrix': self.confusion_matrix}                
+                self.print_results()
+                return self.overview
+
+        def print_results(self):    
+                print('Training dataset size: '+str(len(self.data.index)))
+                print('Training dataset percentage: '+str(100*self.test_size)+"%")            
+                print('_____________________')                
+                print('Confusion-matrix: \n'+str(self.overview['Confusion-matrix']))
+                print('_____________________')
+                print('F-score: '+str(self.overview['F-score']))
+                print('_____________________')
+                print('=====================')

src/backend/data_extractor.py

@@ -0,0 +1,24 @@
+from api import OpenDota, Steam
+import logging
+from context import Context
+from configure_logging import configure_logging
+configure_logging('main.py', logging.INFO)
+
+class DataExtractor:
+        def __init__(self):
+                self.open_dota = OpenDota()
+                self.steam = Steam()
+                self.context = Context()
+
+        def import_matches(self,amount):
+                match_ids = self.open_dota.get_matches_ids(amount)
+                match_collection = self.context.db.get_collection('Matches')
+                for match_id in match_ids:
+                        if match_collection.count_documents({ '_id': match_id }, limit = 1) == 0:
+                                logging.debug('Accessing Steam to get info about match '+str(match_id))
+                                splited_match = self.steam.get_match_detail(match_id)                
+                                if splited_match != None:
+                                        logging.info('Inserting match '+str(splited_match['_id'])+' on database.')
+                                        self.context.insert_one(splited_match,'Matches')
+                        else:
+                                logging.warn('Match '+str(match_id)+' already on database.')

src/backend/knn.py

@@ -0,0 +1,18 @@
+from tables_manager import Tables
+from classifier import Classifier
+from sklearn.neighbors import KNeighborsClassifier
+
+class Knn(Classifier):]
+        def __init__(self,k):
+                super().__init__()
+                self.k = k
+                self.classifier = KNeighborsClassifier(n_neighbors=self.k)
+                self.train()
+                self.results()
+
+        def print_results(self):
+                print('=====================')
+                print('Algorithm: knn')
+                print('K= '+str(self.k))                
+                super().print_results()
+

src/backend/tables_manager.py

@@ -0,0 +1,20 @@
+import numpy as np
+import pandas as pd
+import logging
+import pymongo
+from context import Context
+import json
+
+class Tables:
+        def __init__(self):
+                self.context  = Context()                
+
+        def get_all_matches(self):
+                matches_collection = self.context.db['Matches']
+                matches_cursor = matches_collection.find({})
+                p = list(matches_cursor)
+                result = pd.DataFrame(p)
+                #result['teams'] = result['teams'].map(lambda x :np.concatenate((x[0],x[1])))
+                #result['radiant_win'] = result['radiant_win'] * 1                
+                return result               
+