def plot_crimes_by_coordinates():
category = "DRUNKENNESS"
data = DataDAO.get_data_from_csv('train.csv')
# x=numpy.array([])
# y=numpy.array([])
x = []
y = []
sf_coordinates_x = [-122.6, -122.35]
sf_coordinates_y = [35, 40 ]
for ind,raw in enumerate(data):
if raw[0] != 'Dates' and raw[1] == category:
coord = DataDAO.get_coordinates_tr(raw)
if sf_coordinates_x[0] < coord[0] < sf_coordinates_x[1] and sf_coordinates_y[0] < coord[1] < sf_coordinates_y[1]:
x.append(coord[0])
y.append(coord[1])
# numpy.append(x,[coord[0]])
# numpy.append(y,[coord[1]])
# if ind > 100000:
# break
#heat map
# heatmap, xedges, yedges = numpy.histogram2d(x, y, bins=50)
# extent = [xedges[0], xedges[-1], yedges[0], yedges[-1]]
# plt.subplot(2,1,2)
# plt.imshow(heatmap, extent=extent)
#scatter plot
# plt.subplot(2,1,1)
plt.scatter(x,y,s=1)
plt.show()
def get_test_data(limit=0):
matrix = []
test_data = DataDAO.get_test_vector()
for ind, vector in enumerate(test_data):
if limit != 0 and ind >= limit:
break
matrix.append(vector)
# sparse = csr_matrix(scaled_matrix)
return matrix
def get_test_data(limit=0):
matrix = []
test_data = DataDAO.get_test_vector()
for ind, vector in enumerate(test_data):
if limit != 0 and ind >= limit:
break
matrix.append(vector)
# sparse = csr_matrix(scaled_matrix)
return matrix
def build_sparse_matrix_target(limit=0):
targets_vector = DataDAO.get_targets()
targets_matrix = []
for ind, item in enumerate(targets_vector):
if limit != 0 and ind >= limit:
break
multiclass_vector = TrainingFactory.get_category_vector(item)
targets_matrix.append(multiclass_vector)
sparse = csr_matrix(targets_matrix)
return sparse
def build_target_vector_by_category(category, limit=0):
targets_vector = DataDAO.get_targets()
targets_cat = []
for ind, item in enumerate(targets_vector):
if limit != 0 and ind >= limit:
break
if item == category:
targets_cat.append(1)
else:
targets_cat.append(0)
return targets_cat
def build_sparse_matrix_input(limit=0):
matrix = []
datos = DataDAO.get_train_vector()
for ind, vector in enumerate(datos):
if limit != 0 and ind >= limit:
break
matrix.append(vector)
scaled_matrix = scale(matrix, axis=0)
sparse = csr_matrix(scaled_matrix)
return sparse
def build_sparse_matrix_input(limit=0):
matrix = []
datos = DataDAO.get_train_vector()
for ind, vector in enumerate(datos):
if limit != 0 and ind >= limit:
break
matrix.append(vector)
#center data for 0 mean and sd
scaled_matrix = scale(matrix)
sparse = csr_matrix(scaled_matrix)
# input csrm
return sparse
def plot_crime_category_by_date_time(category):
datos = DataDAO.get_time_features_by_category(category)
dia_year = datos[:,0]
segundo_del_dia = datos[:,1]
dia_semana = datos[:,2]
plt.figure()
plt.suptitle(category)
plt.subplot(3,1,1)
plt.hist(dia_semana, bins=7)
plt.title("Crime by day of week")
plt.subplot(3,1,2)
plt.hist(dia_year, bins=12)
plt.title("Crime by month")
plt.subplot(3,1,3)
plt.hist(segundo_del_dia, bins=24)
plt.title("Crime by hour")
plt.show()
return plt
def get_training_data_by_category(category, limit=0):
limit_pos = limit*0.2
limit_neg = limit*0.8
N_pos = DataDAO.count_training_data_by_category(category)
if N_pos < limit_pos:
limit_pos = N_pos
limit_neg = N_pos*5
training_data = []
training_target = []
positive = DataDAO.get_training_data_by_category(category)
for ind, sample in enumerate(positive):
if limit != 0 and ind >= limit_pos:
break
training_data.append(sample)
training_target.append(1)
negative = DataDAO.get_training_data_by_other_categories(category)
for ind, sample in enumerate(negative):
if limit != 0 and ind >= limit_neg:
break
training_data.append(sample)
training_target.append(0)
scaler = MinMaxScaler()
training_data_scaled = scaler.fit_transform(training_data)
# training_data_scaled = scale(training_data,axis=0)
tr_data_sparse = csr_matrix(training_data_scaled)
return tr_data_sparse, training_target, scaler
