def test_basic(self, Xy_dummy):
eps = 1e-4
X, _, _, _ = Xy_dummy
n_samples, n_channels, lookback, n_assets = X.shape
dtype = X.dtype
device = X.device
network = LinearNet(n_channels, lookback, n_assets)
network.to(device=device, dtype=dtype)
with pytest.raises(ValueError):
network(
torch.ones(n_samples,
n_channels + 1,
lookback,
n_assets,
device=device,
dtype=dtype))
weights = network(X)
assert isinstance(network.hparams, dict)
assert network.hparams
assert torch.is_tensor(weights)
assert weights.shape == (n_samples, n_assets)
assert X.device == weights.device
assert X.dtype == weights.dtype
assert torch.allclose(weights.sum(dim=1),
torch.ones(n_samples).to(dtype=dtype,
device=device),
atol=eps)
def test_n_params(self, n_channels, lookback, n_assets):
network = LinearNet(n_channels, lookback, n_assets)
n_features = n_channels * lookback * n_assets
expected = 0
expected += n_features * 2 # batch norm
expected += n_features * n_assets + n_assets # dense
expected += 1 # temperature
actual = sum(p.numel() for p in network.parameters() if p.requires_grad)
assert expected == actual
n_timesteps = len(data)
# Create features and targets
X_list, y_list = [], []
for i in range(lookback, n_timesteps - horizon - gap + 1):
X_list.append(data[i - lookback:i, :])
y_list.append(data[i + gap:i + gap + horizon, :])
X = np.stack(X_list, axis=0)[:, None, ...]
y = np.stack(y_list, axis=0)[:, None, ...]
# Setup deepdow framework
dataset = InRAMDataset(X, y)
network = LinearNet(1, lookback, n_assets, p=0.5)
dataloader = RigidDataLoader(dataset,
indices=list(range(5000)),
batch_size=batch_size,
lookback=lookback)
val_dataloaders = {
'train':
dataloader,
'val':
RigidDataLoader(dataset,
indices=list(range(5020, 9800)),
batch_size=batch_size,
lookback=lookback)
}
run = Run(network,