def test_eegnet_v4():
rng = np.random.RandomState(42)
n_channels = 18
n_in_times = 500
n_classes = 2
n_samples = 7
X = rng.randn(n_samples, n_channels, n_in_times, 1)
X = torch.Tensor(X.astype(np.float32))
model = EEGNetv4(n_channels, n_classes, input_window_samples=n_in_times)
y_pred = model(X)
assert y_pred.shape == (n_samples, n_classes)
def test_eegnet_v4(input_sizes):
model = EEGNetv4(input_sizes['n_channels'],
input_sizes['n_classes'],
input_window_samples=input_sizes['n_in_times'])
check_forward_pass(model, input_sizes)
batch_size = 32
n_epochs = 200
#one_window.shape : (208, 500)
# Extract number of chans and time steps from dataset
one_window = next(iter(train_loader))[0]
n_chans = one_window.shape[1]
input_window_samples = one_window.shape[2]
model_name = 'deepnet_da'
if model_name == 'eegnet':
#print('Here')
net = EEGNetv4(n_chans,
class_number,
input_window_samples=input_window_samples,
final_conv_length='auto',
drop_prob=0.5)
elif model_name == 'shallowFBCSPnet':
net = ShallowFBCSPNet(
n_chans,
class_number,
input_window_samples=input_window_samples,
final_conv_length='auto',
) # 51%
elif model_name == 'deepnet':
net = deepnet(n_chans, class_number, wind) # 81%
elif model_name == 'deepnet_changeDepth':
net = deepnet_changeDepth(n_chans, class_number, wind, depth) # 81%
model_name = 'deepnet_changeDepth_' + str(depth)
elif model_name == 'deepnet2':