# plot gamma, alpha, accuracy
# save png & 3d data or not
saveDataBool = 0
dir_n = directory + 'res_230319/'
plot_gamma_motor(subject_name, subjects_dir, dir_n, elec_up, mlabViewX,
mlabViewY, saveDataBool)
file_n = directory + 'res_touch.mat'
plot_gamma_touch(subject_name, subjects_dir, file_n, elec_up, mlabViewX,
mlabViewY, saveDataBool)
plot_alpha_motor(subject_name, subjects_dir, dir_n, elec_up, mlabViewX,
mlabViewY, saveDataBool)
plot_alpha_touch(subject_name, subjects_dir, file_n, elec_up, mlabViewX,
mlabViewY, saveDataBool)
dir_n = directory + 'pet67_paper/'
plot_accuracy(subject_name, subjects_dir, dir_n, elec_up, mlabViewX, mlabViewY,
saveDataBool)
# TPS
# source control points
x, y = np.linspace(0, 0.021, 8), np.linspace(0, 0.021,
8) # construct electrodes grid
x, y = np.meshgrid(x, y)
xs = x.flatten()
ys = y.flatten()
control_points = np.vstack([xs, ys, np.zeros(64)]).T
# make T
T = make_T(control_points)
# target control points
points = deepcopy(elec_up) / 1000
# print model structure diagram
print(classifier.summary())
# Compiling the CNN
classifier.compile(adam(lr=.0001),
loss='categorical_crossentropy',
metrics=['accuracy'])
cp_callback = create_callback(checkpoint_path, 1)
hist = classifier.fit_generator(
train_generator,
epochs=num_epoch,
steps_per_epoch=num_train_samples // batch_size,
validation_data=validation_generator,
validation_steps=num_validate_samples // batch_size,
class_weight='auto',
callbacks=[cp_callback])
# Plot training and validation accuracy
plot_accuracy(hist)
# Plot training and validation loss
plot_loss(hist)
# Print test set accuracy and loss values
scores = classifier.evaluate_generator(test_generator,
num_test_samples / batch_size)
print("loss: {}, accuracy: {}".format(scores[0], scores[1]))
'accuracy SD')].sort_values('accuracy MEAN',
ascending=False)).head())
print()
print('Lowest accuracy:')
print((evaluation.loc[(slice(None), slice(None), 'cross-validation',
'Classifier'),
('accuracy MEAN',
'accuracy SD')].sort_values('accuracy MEAN',
ascending=True)).head())
# %% Compare the different classifiers
filenames = ['knn', 'ridge', 'dt']
if True:
# For cross-validation scatter-plot accuracy mean and standard deviation
functions.plot_accuracy(path, filenames)
if True:
# For cross-validation scatter-plot fit time mean and score time
functions.plot_efficiency(path, filenames)
if True:
# List variants of each classifier with highest accuracy values
all_evaluations = pd.DataFrame()
for filename in filenames:
evaluation = pd.read_hdf(os.path.join(path,
filename + '_evaluation.h5'),
key='evaluation')
# Select only rows with cross-validation and only Classifier (no baselines)
rows = ()
for name in evaluation.index.names:
# print model structure diagram
print(model.summary())
# Transfer Learning
print("\nPerforming Transfer Learning")
# Compiling the model
model.compile(optimizer='adam',
loss='categorical_crossentropy',
metrics=['accuracy'])
# Fit the Transfer Learning model to the data from the generators
history = model.fit_generator(train_generator,
epochs=num_epoch,
steps_per_epoch=num_train_samples // batch_size,
validation_data=validation_generator,
validation_steps=num_validate_samples //
batch_size,
class_weight='auto',
shuffle=True)
# Plot training and validation accuracy
plot_accuracy(history)
# Plot training and validation loss
plot_loss(history)
# Print test set accuracy and loss values
scores = model.evaluate_generator(test_generator,
num_test_samples / batch_size)
print("loss: {}, accuracy: {}".format(scores[0], scores[1]))
print('Highest accuracy:')
print((evaluation.loc[(slice(None), slice(None),
'cross-validation', 'Classifier'), ('accuracy MEAN', 'accuracy SD')].
sort_values('accuracy MEAN', ascending=False)).head())
print()
print('Lowest accuracy:')
print((evaluation.loc[(slice(None), slice(None),
'cross-validation', 'Classifier'), ('accuracy MEAN', 'accuracy SD')].
sort_values('accuracy MEAN', ascending=True)).head())
#%% Compare the different classifiers
filenames = ['knn', 'ridge', 'dt']
if False:
# For cross-validation scatter-plot accuracy mean and standard deviation
functions.plot_accuracy(cfg.default.occupancy_figures, filenames)
if False:
# For cross-validation scatter-plot fit time mean and score time
functions.plot_efficiency(cfg.default.occupancy_figures, filenames)
if False:
# List variants of each classifier with highest accuracy values
all_evaluations = pd.DataFrame()
for filename in filenames:
path = cfg.default.occupancy_figures
evaluation = pd.read_hdf(os.path.join(path, filename + '_evaluation.h5'),
key='evaluation')
# Select only rows with cross-validation and only Classifier (no baselines)
rows = ()
for name in evaluation.index.names: