13    os.system("rm -rf /home/linuxsagar/PycharmProjects/FruitClassifierTesting/FCTesting/static/histo.png")
14
15
16def make_histogram(i_hist, i_hufe, i_ehfe, i_hhfe, dataDir):17    fig, axes = plt.subplots(nrows=4, ncols=4)
18    fig.tight_layout()
19

 9    os.mkdir('/home/linuxsagar/tempTest/')
10
11
12def clean_graph_dir():13    os.system("rm -rf /home/linuxsagar/PycharmProjects/FruitClassifierTesting/FCTesting/static/histo.png")
14
15

 4from matplotlib.pyplot import plot
 5
 6
 7def clean_dir(): 8    shutil.rmtree('/home/linuxsagar/tempTest/')
 9    os.mkdir('/home/linuxsagar/tempTest/')
10

 2import os
 3
 4
 5def clean_dir(): 6    shutil.rmtree('/home/linuxsagar/tempTest/')
 7    os.mkdir('/home/linuxsagar/tempTest/')
 8

 7    os.mkdir('/home/linuxsagar/tempTest/')
 8
 9
10def clean_graph_dir():11    # shutil.rmtree('/home/linuxsagar/PycharmProjects/FruitClassifierTesting/graphs/')
12    # os.mkdir('')
13    # os.remove()

24labels = pickle.load(open(os.path.join(dataDir, "../data/featLabel.pkl"), 'rb'))
25
26
27def predictor():28    for img in glob('/home/linuxsagar/tempTest/*'):
29
30        # List to hold the feature of new Image

12app.config['ALLOWED_EXTENSIONS'] = {'png', 'jpg', 'jpeg'}
13
14
15def allowed_file(filename):16    return '.' in filename and filename.rsplit('.', 1)[1] in app.config['ALLOWED_EXTENSIONS']
17
18

17
18
19@app.route('/')
20def index():21    MainServices.clean_dir()
22    MainServices.clean_graph_dir()
23    return render_template('index.html')

24
25
26@app.route('/upload', methods=['POST'])
27def upload():28    MainServices.clean_graph_dir()
29    file = request.files['file']
30    if file and allowed_file(file.filename):

34
35
36@app.route('/predict')
37def predict():38    data = predictor()
39    MainServices.clean_dir()
40    return render_template('index.html', result=data[0], hue=data[1], edge=data[2], haar=data[3])

41
42
43@app.errorhandler(Exception)
44def all_exception_handler(error):45    return render_template('error.html')
46
47

23
24
25# Extract features and target labels for training
26def get_feature_labels():27    features = []
28    labels = []
29    for im in images:

35    return [features, labels]
36
37
38def frange(start, stop, step):39    i = start
40    while i < stop:
41        yield i

46
47
48# Method to choose best learning rate
49def choose_learning_rate():50    for i in frange(0.1, 1.0, 0.1):
51        rf = AdaBoostClassifier(n_estimators=100, learning_rate=i)
52        clf = Pipeline(steps=[('preprocess', StandardScaler()), ('clasifier', rf)])

59
60
61# Training function
62def training():63    pickle.dump(feat_label, open('featLabel.pkl', 'wb'))
64    adb = AdaBoostClassifier(n_estimators=100, learning_rate=0.7)
65    clf = Pipeline(steps=[('preprocess', StandardScaler()), ('clasifier', adb)])

73
74
75# System testing
76def test_system():77    # path of test folder
78    with open("classifier.pkl", "rb") as file:
79        for each in glob('./testing/*'):