shuffle=True)
return trainloader, dim, number_rows_data
# -
def large_cycle(map_, training_data):
basic_visualization(map_display(map_.map))
print(map_display(map_.map))
for i in range(number_iterations):
cycle(map_, training_data)
basic_visualization(map_display(map_.map))
print(map_display(map_.map))
training, dim, number_rows_data = load_data(data)
map1 = MapClass(length, width, dim, move_closer_coef, number_iterations)
map1.weights_to_map()
map1.step(training, verbose=True)
map1.cycle(training, verbose=True)
map1.classify_all(map1.convert_data_tensor(data))
map1.convert_data_tensor(data)
if dim == 1:
return map_.view(length, width)
else:
return map_.view(dim, length, width)
def large_cycle(map_, training_data):
basic_visualization(map_display(map_.map))
print(map_display(map_.map))
for i in range(number_iterations):
cycle(map_, training_data)
basic_visualization(map_display(map_.map))
print(map_display(map_.map))
training = load_data(data)
map1 = MapClass(length, width, dim, move_closer_coef)
map1.map
map1.cycle(training)
map1.map
map1.distance_matrix
map1.impact_matrix
basic_visualization(map1.map)
# +
# Network configuration
data = rgb_colors
data_lables = color_names
batch_size = 2
length = 10
width = 10
number_epochs = 100
shuffle = True
learning_rate = 0.01
# -
map1 = MapClass(data, length, width, learning_rate, number_epochs, matrix1,
data_lables, batch_size, shuffle)
# +
# map1.weights
# +
# training, dim, number_rows_data = load_data(data, batch_size)
# -
plt.rcParams['figure.dpi'] = 150
map1.large_cycle(draw_every_epoch=10, rgb=True)
map1.cycle()
