def test_eegnet_v1(input_sizes):
model = EEGNetv1(input_sizes['n_channels'],
input_sizes['n_classes'],
input_window_samples=input_sizes['n_in_times'])
check_forward_pass(
model,
input_sizes,
)
def test_eegnet_v1():
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 = EEGNetv1(n_channels, n_classes, input_window_samples=n_in_times)
y_pred = model(X)
assert y_pred.shape == (n_samples, n_classes)
cuda = torch.cuda.is_available(
) # check if GPU is available, if True chooses to use it
device = 'cuda' if cuda else 'cpu'
if cuda:
torch.backends.cudnn.benchmark = True
seed = 20200220 # random seed to make results reproducible
# Set random seed to be able to reproduce results
set_random_seeds(seed=seed, cuda=cuda)
n_classes = 3
# Extract number of chans and time steps from dataset
n_chans = train_set[0][0].shape[0]
input_window_samples = train_set[0][0].shape[1]
#model = ShallowFBCSPNet(n_chans,n_classes,input_window_samples=input_window_samples,final_conv_length='auto',)
model = EEGNetv1(n_chans, n_classes, input_window_samples)
# Send model to GPU
if cuda:
model.cuda()
#x = torch.randn(1, n_chans, input_window_samples) # input: torch.Size([64, 64, 500])
#y=model(x)
# These values we found good for shallow network:
lr = 0.0625 * 0.01
weight_decay = 0
batch_size = 64
n_epochs = 4
clf = EEGClassifier(
model,