def conf_matrix(self, y_test, y_pred, clf, uniqueNames):
cnf_matrix = confusion_matrix(y_test.values, y_pred)
np.set_printoptions(precision=2)
# Plot non-normalized confusion matrix
plt.figure()
vis = Visualisation()
class_names = uniqueNames
vis.plot_confusion_matrix(
clf,
cnf_matrix,
classes=class_names,
title='Confusion matrix, without normalization')
def __init__(self, configFilepath=None):
#configFilepath = r"C:\Users\ashvin\Desktop\UnderDevelopment\sentimentConfig.csv"
df = pd.read_csv(configFilepath)
df.set_index('Parameters', inplace=True)
reviews = df.loc['paths', 'Value'].split(',')
catNum = df.loc['numOfCat', 'Value']
catTypes = df.loc['catType', 'Value'].split(',')
sentimentLabel = df.loc['train', 'Value']
sentiment = sa.SentimentTrain()
acc, allReviews = sentiment.extract(reviews, catNum, catTypes,
sentimentLabel)
visuals = Visualisation(df=None, target=None)
visuals.wordCloud(allReviews)
def __init__(self, df, clusterSizes=None):
Unsupervised.__init__(self)
self.df = df
if clusterSizes == None:
self.scores = []
clusterSizes = [2, 3, 4, 5, 6, 7, 8, 9, 10]
for clusterSize in clusterSizes:
self.kmeans = KMeans(n_clusters=clusterSize)
self.fit()
self.labels, self.centroids, self.silhouette_avg = self.predict(
)
self.scores.append(self.silhouette_avg)
visuals = Visualisation()
visuals.silhouetteScores(self.scores, clusterSizes)
else:
self.kmeans = KMeans(n_clusters=clusterSizes)
self.fit()
self.labels, self.centroids, _ = self.predict()
def __init__(self, df=None):
#configFilepath = r"C:\Users\ashvin\Desktop\UnderDevelopment\sentimentConfig.csv"
#df = pd.read_csv(configFilepath)
#df.set_index('Parameters', inplace = True)
print('creating Results directory--------> at {}'.format(os.getcwd()))
try:
os.makedirs('results')
except:
pass
reviews = df.loc['paths', 'Value'].split(',')
catNum = df.loc['numOfCat', 'Value']
catTypes = df.loc['catType', 'Value'].split(',')
sentimentLabel = df.loc['train', 'Value']
sentiment = sa.SentimentTrain()
acc, allReviews = sentiment.extract(reviews, catNum, catTypes,
sentimentLabel)
visuals = Visualisation(df=None, target=None)
visuals.wordCloud(allReviews)
def __init__(self, configFilepath=None):
self.configFilepath = configFilepath
configDf = pd.read_csv(configFilepath, sep=',')
configDf.set_index('Parameters', inplace=True)
columnsConsidered = configDf.loc['ColumnsConsidered', 'Value']
imputation = configDf.loc['Imputation', 'Value']
target = configDf.loc['Target', 'Value']
clf = configDf.loc['Classifier', 'Value']
modelingType = configDf.loc['ModelingType', 'Value']
filepath = configDf.loc['filepath', 'Value']
separator = configDf.loc['separator', 'Value']
temp1 = DataQualityCheck(filepath,
target=target,
separator=separator,
columnsConsidered=columnsConsidered)
df1 = temp1.considerColumns()
temp1.checkMissingValues()
temp1.checkOutliers()
temp2 = DataPreparation(imputation=imputation)
df1 = temp2.convertCategoricalToDummy(df1)
df1 = temp2.imputation(df1)
df1 = temp2.featureNormalisation(df1)
if modelingType == 'classification':
if clf == 'Logistic':
temp3 = Logistic(target=target, df=df1)
scores = temp3.runLogistic()
elif clf == 'SVM':
temp3 = SVM(target=target, df=df1)
scores = temp3.runSVM()
elif clf == 'RF':
temp3 = RandomForestClf(target=target, df=df1)
scores = temp3.runRF()
elif clf == 'NN':
temp3 = NeuralNetwork(target=target, df=df1)
scores = temp3.runNN()
elif modelingType == 'regression':
if clf == 'Linear':
temp3 = Linear(target=target, df=df1)
scores = temp3.runLinear()
elif clf == 'LassoRegression':
temp3 = LassoRegression(target=target, df=df1)
scores = temp3.runLasso()
elif clf == 'RidgeRegression':
temp3 = RidgeRegression(target=target, df=df1)
scores = temp3.runRidge()
elif clf == 'RandomForest':
temp3 = RandomForestReg(target=target, df=df1)
scores = temp3.runRF()
print('scores for classifier {} are: {}'.format(
configDf.loc['Classifier', 'Value'], scores))
temp4 = Visualisation(df=df1, target=target)
agents = [LeftAgent, StayAgent, RandomAgent]
world = World(WORLD_SIZE)
world.add_obstacles(COUNT_OBSTACLES)
world.add_food(COUNT_FOOD)
for player_index in range(COUNT_PLAYERS):
random_position = [np.random.randint(0, WORLD_SIZE[0]), np.random.randint(0, WORLD_SIZE[1])]
bot = Player(random_position, player_index)
Agent = np.random.choice(agents)
bot.set_agent(Agent())
world.player_register(bot)
if IS_VISUALISATION:
view = Visualisation(world)
running = True
counter = 1
while running and counter < COUNT_EPOCH:
world.tick()
if IS_VISUALISATION:
view.draw_objects()
view.draw_players()
view.append_food()
view.vis_game.display.flip()
for event in view.vis_game.event.get():
if event.type == view.vis_game.QUIT:
running = False
else:
print(world)
def ROCplots(self, ytest, ypred_prob, clf):
vis = Visualisation()
vis.ROCplots(ytest, ypred_prob, clf)
def residVsFitted(self, ypred, ytest, clf):
vis = Visualisation()
vis.residVsPredPlots(ypred, ytest, clf)
def normalQQ(self, ypred, ytest, clf):
vis = Visualisation()
vis.normalQQplots(ypred, ytest, clf)
def precisionVsrecallPlots(self, ytest, ypred_prob, clf):
vis = Visualisation()
vis.precVsRecall(ytest, ypred_prob, clf)
def jointPlot(self, df, target):
vis = Visualisation()
vis.snsJointPlots(df, target)
def saveBoxPlots(self, df):
vis = Visualisation()
vis.boxplot(df)
def savePairPlots(self, df):
vis = Visualisation()
vis.pairplot(df)
def hist(self, df):
vis = Visualisation()
vis.histogram(df)
def corr_plot(self, df, clf):
vis = Visualisation()
vis.correlationPlot(df, clf)
def __init__(self):
self.visualization = Visualisation()
self.controller = Controller()
def swarmPlots(self, df, target):
vis = Visualisation()
vis.snsSwarmPlots(df, target)