autora.theorist.darts.architect

Architect

Bases: object

A learner operating on the architecture weights of a DARTS model. This learner handles training the weights associated with mixture operations (architecture weights).

Source code in temp_dir/darts/src/autora/theorist/darts/architect.py

class Architect(object):
    """
    A learner operating on the architecture weights of a DARTS model.
    This learner handles training the weights associated with mixture operations
    (architecture weights).
    """

    def __init__(
        self,
        model: Network,
        arch_learning_rate_max: float,
        arch_momentum: float,
        arch_weight_decay: float,
        arch_weight_decay_df: float = 0,
        arch_weight_decay_base: float = 0,
        fair_darts_loss_weight: float = 1,
    ):
        """
        Initializes the architecture learner.

        Arguments:
            model: a network model implementing the full DARTS model.
            arch_learning_rate_max: learning rate for the architecture weights
            arch_momentum: arch_momentum used in the Adam optimizer for architecture weights
            arch_weight_decay: general weight decay for the architecture weights
            arch_weight_decay_df: (weight decay applied to architecture weights in proportion
                to the number of parameters of an operation)
            arch_weight_decay_base: (a constant weight decay applied to architecture weights)
            fair_darts_loss_weight: (a regularizer that pushes architecture weights more toward
                zero or one in the fair DARTS variant)
        """
        # set parameters for architecture learning
        self.network_arch_momentum = arch_momentum
        self.network_weight_decay = arch_weight_decay
        self.network_weight_decay_df = arch_weight_decay_df
        self.arch_weight_decay_base = arch_weight_decay_base * model._steps
        self.fair_darts_loss_weight = fair_darts_loss_weight

        self.model = model
        self.lr = arch_learning_rate_max
        # architecture is optimized using Adam
        self.optimizer = torch.optim.Adam(
            self.model.arch_parameters(),
            lr=arch_learning_rate_max,
            betas=(0.5, 0.999),
            weight_decay=arch_weight_decay,
        )

        # initialize weight decay matrix
        self._init_decay_weights()

        # initialize the logged loss
        self.current_loss = 0

    def _init_decay_weights(self):
        """
        This function initializes the weight decay matrix. The weight decay matrix
        is subtracted from the architecture weight matrix on every learning step. The matrix
        specifies a weight decay which is proportional to the number of parameters used in an
        operation.
        """
        n_params = list()
        for operation in self.model.cells._ops[0]._ops:
            if isiterable(operation):
                n_params_total = (
                    1  # any non-zero operation is counted as an additional parameter
                )
                for subop in operation:
                    for parameter in subop.parameters():
                        if parameter.requires_grad is True:
                            n_params_total += parameter.data.numel()
            else:
                n_params_total = 0  # no operation gets zero parameters
            n_params.append(n_params_total)

        self.decay_weights = Variable(
            torch.zeros(self.model.arch_parameters()[0].data.shape)
        )
        for idx, param in enumerate(n_params):
            if param > 0:
                self.decay_weights[:, idx] = (
                    param * self.network_weight_decay_df + self.arch_weight_decay_base
                )
            else:
                self.decay_weights[:, idx] = param
        self.decay_weights = self.decay_weights
        self.decay_weights = self.decay_weights.data

    def _compute_unrolled_model(
        self,
        input: torch.Tensor,
        target: torch.Tensor,
        eta: float,
        network_optimizer: torch.optim.Optimizer,
    ):
        """
        Helper function used to compute the approximate architecture gradient.

        Arguments:
            input: input patterns
            target: target patterns
            eta: learning rate
            network_optimizer: optimizer used to updating the architecture weights

        Returns:
            unrolled_model: the unrolled architecture
        """
        loss = self.model._loss(input, target)
        theta = _concat(self.model.parameters()).data
        try:
            moment = _concat(
                network_optimizer.state[v]["momentum_buffer"]
                for v in self.model.parameters()
            ).mul_(self.network_arch_momentum)
        except Exception:
            moment = torch.zeros_like(theta)
        dtheta = (
            _concat(torch.autograd.grad(loss, self.model.parameters())).data
            + self.network_weight_decay * theta
        )
        unrolled_model = self._construct_model_from_theta(
            theta.sub(eta, moment + dtheta)
        )
        return unrolled_model

    def step(
        self,
        input_valid: torch.Tensor,
        target_valid: torch.Tensor,
        network_optimizer: torch.optim.Optimizer,
        unrolled: bool,
        input_train: Optional[torch.Tensor] = None,
        target_train: Optional[torch.Tensor] = None,
        eta: float = 1,
    ):
        """
        Updates the architecture parameters for one training iteration

        Arguments:
            input_valid: input patterns for validation set
            target_valid: target patterns for validation set
            network_optimizer: optimizer used to updating the architecture weights
            unrolled: whether to use the unrolled architecture or not (i.e., whether to use
                the approximate architecture gradient or not)
            input_train: input patterns for training set
            target_train: target patterns for training set
            eta: learning rate for the architecture weights
        """

        # input_train, target_train only needed for approximation (unrolled=True)
        # of architecture gradient
        # when performing a single weigh update

        # initialize gradients to be zero
        self.optimizer.zero_grad()
        # use different backward step depending on whether to use
        # 2nd order approximation for gradient update
        if unrolled:  # probably using eta of parameter update here
            self._backward_step_unrolled(
                input_train,
                target_train,
                input_valid,
                target_valid,
                eta,
                network_optimizer,
            )
        else:
            self._backward_step(input_valid, target_valid)
        # move Adam one step
        self.optimizer.step()

    # backward step (using non-approximate architecture gradient, i.e., full training)
    def _backward_step(self, input_valid: torch.Tensor, target_valid: torch.Tensor):
        """
        Computes the loss and updates the architecture weights assuming full optimization
        of coefficients for the current architecture.

        Arguments:
            input_valid: input patterns for validation set
            target_valid: target patterns for validation set
        """
        if self.model.DARTS_type == DARTSType.ORIGINAL:
            loss = self.model._loss(input_valid, target_valid)
        elif self.model.DARTS_type == DARTSType.FAIR:
            loss1 = self.model._loss(input_valid, target_valid)
            loss2 = -F.mse_loss(
                torch.sigmoid(self.model.alphas_normal),
                0.5 * torch.ones(self.model.alphas_normal.shape, requires_grad=False),
            )  # torch.tensor(0.5, requires_grad=False)
            loss = loss1 + self.fair_darts_loss_weight * loss2
        else:
            raise Exception(
                "DARTS Type " + str(self.model.DARTS_type) + " not implemented"
            )

        loss.backward()
        self.current_loss = loss.item()

        # weight decay proportional to degrees of freedom
        for p in self.model.arch_parameters():
            p.data.sub_((self.decay_weights * self.lr))  # weight decay

    # backward pass using second order approximation
    def _backward_step_unrolled(
        self,
        input_train: torch.Tensor,
        target_train: torch.Tensor,
        input_valid: torch.Tensor,
        target_valid: torch.Tensor,
        eta: float,
        network_optimizer: torch.optim.Optimizer,
    ):
        """
        Computes the loss and updates the architecture weights using the approximate architecture
        gradient.

        Arguments:
            input_train: input patterns for training set
            target_train: target patterns for training set
            input_valid: input patterns for validation set
            target_valid: target patterns for validation set
            eta: learning rate
            network_optimizer: optimizer used to updating the architecture weights

        """

        # gets the model
        unrolled_model = self._compute_unrolled_model(
            input_train, target_train, eta, network_optimizer
        )

        if self.model.DARTS_type == DARTSType.ORIGINAL:
            unrolled_loss = unrolled_model._loss(input_valid, target_valid)
        elif self.model.DARTS_type == DARTSType.FAIR:
            loss1 = self.model._loss(input_valid, target_valid)
            loss2 = -F.mse_loss(
                torch.sigmoid(self.model.alphas_normal),
                torch.tensor(0.5, requires_grad=False),
            )
            unrolled_loss = loss1 + self.fair_darts_loss_weight * loss2
        else:
            raise Exception(
                "DARTS Type " + str(self.model.DARTS_type) + " not implemented"
            )

        unrolled_loss.backward()
        dalpha = [v.grad for v in unrolled_model.arch_parameters()]
        vector = [v.grad.data for v in unrolled_model.parameters()]
        implicit_grads = self._hessian_vector_product(vector, input_train, target_train)

        for g, ig in zip(dalpha, implicit_grads):
            g.data.sub_(eta, ig.data)

        for v, g in zip(self.model.arch_parameters(), dalpha):
            if v.grad is None:
                v.grad = Variable(g.data)
            else:
                v.grad.data.copy_(g.data)

    def _construct_model_from_theta(self, theta: torch.Tensor):
        """
        Helper function used to compute the approximate gradient update
        for the architecture weights.

        Arguments:
            theta: term used to compute approximate gradient update

        """
        model_new = self.model.new()
        model_dict = self.model.state_dict()

        params, offset = {}, 0
        for k, v in self.model.named_parameters():
            v_length = np.prod(v.size())
            params[k] = theta[offset : (offset + v_length)].view(v.size())
            offset += v_length

        assert offset == len(theta)
        model_dict.update(params)
        model_new.load_state_dict(model_dict)
        return model_new  # .cuda() # Edit SM 10/26/19: uncommented for cuda

    # second order approximation of architecture gradient (see Eqn. 8 from Liu et al, 2019)
    def _hessian_vector_product(
        self, vector: torch.Tensor, input: torch.Tensor, target: torch.Tensor, r=1e-2
    ):
        """
        Helper function used to compute the approximate gradient update
        for the architecture weights. It computes the hessian vector product outlined in Eqn. 8
        from Liu et al, 2019.

        Arguments:
            vector: input vector
            input: input patterns
            target: target patterns
            r: coefficient used to compute the hessian vector product

        """
        R = r / _concat(vector).norm()
        for p, v in zip(self.model.parameters(), vector):
            p.data.add_(R, v)
        loss = self.model._loss(input, target)
        grads_p = torch.autograd.grad(loss, self.model.arch_parameters())

        for p, v in zip(self.model.parameters(), vector):
            p.data.sub_(2 * R, v)
        loss = self.model._loss(input, target)
        grads_n = torch.autograd.grad(loss, self.model.arch_parameters())

        for p, v in zip(self.model.parameters(), vector):
            p.data.add_(R, v)

        # this implements Eqn. 8 from Liu et al. (2019)
        return [(x - y).div_(2 * R) for x, y in zip(grads_p, grads_n)]

__init__(model, arch_learning_rate_max, arch_momentum, arch_weight_decay, arch_weight_decay_df=0, arch_weight_decay_base=0, fair_darts_loss_weight=1)

Initializes the architecture learner.

Parameters:

Name	Type	Description	Default
model	Network	a network model implementing the full DARTS model.	required
arch_learning_rate_max	float	learning rate for the architecture weights	required
arch_momentum	float	arch_momentum used in the Adam optimizer for architecture weights	required
arch_weight_decay	float	general weight decay for the architecture weights	required
arch_weight_decay_df	float	(weight decay applied to architecture weights in proportion to the number of parameters of an operation)	0
arch_weight_decay_base	float	(a constant weight decay applied to architecture weights)	0
fair_darts_loss_weight	float	(a regularizer that pushes architecture weights more toward zero or one in the fair DARTS variant)	1

Source code in temp_dir/darts/src/autora/theorist/darts/architect.py

def __init__(
    self,
    model: Network,
    arch_learning_rate_max: float,
    arch_momentum: float,
    arch_weight_decay: float,
    arch_weight_decay_df: float = 0,
    arch_weight_decay_base: float = 0,
    fair_darts_loss_weight: float = 1,
):
    """
    Initializes the architecture learner.

    Arguments:
        model: a network model implementing the full DARTS model.
        arch_learning_rate_max: learning rate for the architecture weights
        arch_momentum: arch_momentum used in the Adam optimizer for architecture weights
        arch_weight_decay: general weight decay for the architecture weights
        arch_weight_decay_df: (weight decay applied to architecture weights in proportion
            to the number of parameters of an operation)
        arch_weight_decay_base: (a constant weight decay applied to architecture weights)
        fair_darts_loss_weight: (a regularizer that pushes architecture weights more toward
            zero or one in the fair DARTS variant)
    """
    # set parameters for architecture learning
    self.network_arch_momentum = arch_momentum
    self.network_weight_decay = arch_weight_decay
    self.network_weight_decay_df = arch_weight_decay_df
    self.arch_weight_decay_base = arch_weight_decay_base * model._steps
    self.fair_darts_loss_weight = fair_darts_loss_weight

    self.model = model
    self.lr = arch_learning_rate_max
    # architecture is optimized using Adam
    self.optimizer = torch.optim.Adam(
        self.model.arch_parameters(),
        lr=arch_learning_rate_max,
        betas=(0.5, 0.999),
        weight_decay=arch_weight_decay,
    )

    # initialize weight decay matrix
    self._init_decay_weights()

    # initialize the logged loss
    self.current_loss = 0

step(input_valid, target_valid, network_optimizer, unrolled, input_train=None, target_train=None, eta=1)

Updates the architecture parameters for one training iteration

Parameters:

Name	Type	Description	Default
input_valid	Tensor	input patterns for validation set	required
target_valid	Tensor	target patterns for validation set	required
network_optimizer	Optimizer	optimizer used to updating the architecture weights	required
unrolled	bool	whether to use the unrolled architecture or not (i.e., whether to use the approximate architecture gradient or not)	required
input_train	Optional[Tensor]	input patterns for training set	None
target_train	Optional[Tensor]	target patterns for training set	None
eta	float	learning rate for the architecture weights	1

Source code in temp_dir/darts/src/autora/theorist/darts/architect.py

def step(
    self,
    input_valid: torch.Tensor,
    target_valid: torch.Tensor,
    network_optimizer: torch.optim.Optimizer,
    unrolled: bool,
    input_train: Optional[torch.Tensor] = None,
    target_train: Optional[torch.Tensor] = None,
    eta: float = 1,
):
    """
    Updates the architecture parameters for one training iteration

    Arguments:
        input_valid: input patterns for validation set
        target_valid: target patterns for validation set
        network_optimizer: optimizer used to updating the architecture weights
        unrolled: whether to use the unrolled architecture or not (i.e., whether to use
            the approximate architecture gradient or not)
        input_train: input patterns for training set
        target_train: target patterns for training set
        eta: learning rate for the architecture weights
    """

    # input_train, target_train only needed for approximation (unrolled=True)
    # of architecture gradient
    # when performing a single weigh update

    # initialize gradients to be zero
    self.optimizer.zero_grad()
    # use different backward step depending on whether to use
    # 2nd order approximation for gradient update
    if unrolled:  # probably using eta of parameter update here
        self._backward_step_unrolled(
            input_train,
            target_train,
            input_valid,
            target_valid,
            eta,
            network_optimizer,
        )
    else:
        self._backward_step(input_valid, target_valid)
    # move Adam one step
    self.optimizer.step()