Python nd.sum方法代码示例

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def _spectral_norm(self):
        """ spectral normalization """
        w = self.params.get('weight').data(self.ctx)
        w_mat = nd.reshape(w, [w.shape[0], -1])

        _u = self.u.data(self.ctx)
        _v = None

        for _ in range(POWER_ITERATION):
            _v = nd.L2Normalization(nd.dot(_u, w_mat))
            _u = nd.L2Normalization(nd.dot(_v, w_mat.T))

        sigma = nd.sum(nd.dot(_u, w_mat) * _v)
        if sigma == 0.:
            sigma = EPSILON

        self.params.setattr('u', _u)

        return w / sigma 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def hybrid_forward(self, F, pred, label, sample_weight=None):
        """Loss forward"""
        if not self._from_logits:
            pred = F.sigmoid(pred)
        if self._sparse_label:
            one_hot = F.one_hot(label, self._num_class)
        else:
            one_hot = label > 0
        pt = F.where(one_hot, pred, 1 - pred)
        t = F.ones_like(one_hot)
        alpha = F.where(one_hot, self._alpha * t, (1 - self._alpha) * t)
        loss = -alpha * ((1 - pt) ** self._gamma) * F.log(F.minimum(pt + self._eps, 1))
        loss = _apply_weighting(F, loss, self._weight, sample_weight)
        if self._size_average:
            return F.mean(loss, axis=self._batch_axis, exclude=True)
        else:
            return F.sum(loss, axis=self._batch_axis, exclude=True) 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def Route(self, x):
        # b_mat = nd.repeat(self.b_mat.data(), repeats=x.shape[0], axis=0)#nd.stop_gradient(nd.repeat(self.b_mat.data(), repeats=x.shape[0], axis=0))
        b_mat = nd.zeros((x.shape[0],1,self.num_cap, self.num_locations), ctx=x.context)
        x_expand = nd.expand_dims(nd.expand_dims(x, axis=2),2)
        w_expand = nd.repeat(nd.expand_dims(self.w_ij.data(x.context),axis=0), repeats=x.shape[0], axis=0)
        u_ = w_expand*x_expand
        # u_ = nd.abs(w_expand - x_expand)
        u = nd.sum(u_, axis = 1)
        u_no_gradient = nd.stop_gradient(u)
        for i in range(self.route_num):
            c_mat = nd.softmax(b_mat, axis=2)
            if i == self.route_num -1:
                s = nd.sum(u * c_mat, axis=-1)
            else:
                s = nd.sum(u_no_gradient * c_mat, axis=-1)
            v = squash(s, 1)
            v1 = nd.expand_dims(v, axis=-1)
            if i != self.route_num - 1:
                update_term = nd.sum(u_no_gradient*v1, axis=1, keepdims=True)
                b_mat = b_mat + update_term
        return v 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def get_distance_matrix(x):
    """Get distance matrix given a matrix. Used in testing."""
    square = nd.sum(x ** 2.0, axis=1, keepdims=True)
    distance_square = square + square.transpose() - (2.0 * nd.dot(x, x.transpose()))
    return nd.sqrt(distance_square) 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def __init__(self, mods, aggregate='sum'):
        super(HeteroGraphConv, self).__init__()
        with self.name_scope():
            for name, mod in mods.items():
                self.register_child(mod, name)
            self.mods = mods
            if isinstance(aggregate, str):
                self.agg_fn = get_aggregate_fn(aggregate)
            else:
                self.agg_fn = aggregate 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def _mixup_forward(self, F, pred, label1, label2, lam, sample_weight=None):
        if not self._from_logits:
            pred = F.log_softmax(pred, self._axis)
        if self._sparse_label:
            loss1 = -F.pick(pred, label1, axis=self._axis, keepdims=True)
            loss2 = -F.pick(pred, label2, axis=self._axis, keepdims=True)
            loss = lam * loss1 + (1 - lam) * loss2
        else:
            label1 = _reshape_like(F, label1, pred)
            label2 = _reshape_like(F, label2, pred)
            loss1 = -F.sum(pred*label1, axis=self._axis, keepdims=True)
            loss2 = -F.sum(pred*label2, axis=self._axis, keepdims=True)
            loss = lam * loss1 + (1 - lam) * loss2
        loss = _apply_weighting(F, loss, self._weight, sample_weight)
        return F.mean(loss, axis=self._batch_axis, exclude=True) 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def weight_l1_loss(self, F, pred_loc, label_loc, loss_weight):
        """Compute weight_l1_loss"""
        pred_loc = pred_loc.reshape((self.b, 4, -1, self.h, self.w))
        diff = F.abs((pred_loc - label_loc))
        diff = F.sum(diff, axis=1).reshape((self.b, -1, self.h, self.w))
        loss = diff * loss_weight
        return F.sum(loss)/self.b 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def evaluate_accuracy(data_iterator, net, ctx=[mx.cpu()]):
    if isinstance(ctx, mx.Context):
        ctx = [ctx]
    acc = nd.array([0])
    n = 0.
    if isinstance(data_iterator, mx.io.MXDataIter):
        data_iterator.reset()
    for batch in data_iterator:
        data, label, batch_size = _get_batch(batch, ctx)
        for X, y in zip(data, label):
            acc += nd.sum(net(X).argmax(axis=1) == y).copyto(mx.cpu())
            n += y.size
        acc.wait_to_read()  # don't push too many operators into backend
    return acc.asscalar() / n 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def train(train_data, test_data, net, loss, trainer, ctx, num_epochs, print_batches=None):
    """Train a network"""
    print("Start training on ", ctx)
    if isinstance(ctx, mx.Context):
        ctx = [ctx]
    for epoch in range(num_epochs):
        train_loss, train_acc, n, m = 0.0, 0.0, 0.0, 0.0
        if isinstance(train_data, mx.io.MXDataIter):
            train_data.reset()
        start = time()
        for i, batch in enumerate(train_data):
            data, label, batch_size = _get_batch(batch, ctx)
            losses = []
            with autograd.record():
                outputs = [net(X) for X in data]
                losses = [loss(yhat, y) for yhat, y in zip(outputs, label)]
            for l in losses:
                l.backward()
            train_acc += sum([(yhat.argmax(axis=1) == y).sum().asscalar()
                              for yhat, y in zip(outputs, label)])
            train_loss += sum([l.sum().asscalar() for l in losses])
            trainer.step(batch_size)
            n += batch_size
            m += sum([y.size for y in label])
            if print_batches and (i + 1) % print_batches == 0:
                print("Batch %d. Loss: %f, Train acc %f" % (
                    n, train_loss / n, train_acc / m
                ))

        test_acc = evaluate_accuracy(test_data, net, ctx)
        print("Epoch %d. Loss: %.3f, Train acc %.2f, Test acc %.2f, Time %.1f sec" % (
            epoch, train_loss / n, train_acc / m, test_acc, time() - start
        )) 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def grad_clipping(params, clipping_norm, ctx):
    """Gradient clipping."""
    if clipping_norm is not None:
        norm = nd.array([0.0], ctx)
        for p in params:
            norm += nd.sum(p.grad ** 2)
        norm = nd.sqrt(norm).asscalar()
        if norm > clipping_norm:
            for p in params:
                p.grad[:] *= clipping_norm / norm 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def squash(x, axis):
    s_squared_norm = nd.sum(nd.square(x), axis, keepdims=True)
    # if s_squared_norm is really small, we will be in trouble
    # so I removed the s_quare terms
    # scale = s_squared_norm / ((1 + s_squared_norm) * nd.sqrt(s_squared_norm + 1e-9))
    # return x * scale
    scale = nd.sqrt(s_squared_norm + 1e-9)
    return x / scale 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def route(self, u):
        b_mat = nd.zeros((u.shape[0], self.num_cap_in, self.num_cap, 1, u.shape[4], u.shape[5]), ctx=u.context)
        for i in range(self.route_num):
            c_mat = nd.softmax(b_mat, axis=2)
            s = nd.sum(u * c_mat, axis=1)
            v = squash(s, 2)
            if i != self.route_num - 1:
                v1 = nd.expand_dims(v, axis=1)
                update_term = nd.sum(u*v1, axis=3, keepdims=True)
                b_mat = b_mat + update_term
        return v 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def forward(self, x):
        x = nd.sqrt(nd.sum(nd.square(x), 1))
        return x 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def update(self, data, label, mask):
		data = self.scaler.inverse_transform(data)
		label = self.scaler.inverse_transform(label)
		
		_cnt = nd.sum(mask).as_in_context(mx.cpu())
		_loss = nd.sum((data - label) ** 2 * mask).as_in_context(mx.cpu())
		if self.cnt is None:
			self.cnt = 0
			self.loss = 0
		self.cnt += _cnt
		self.loss += _loss 

# 需要导入模块: from mxnet import nd [as 别名]
# 或者: from mxnet.nd import sum [as 别名]
def update(self, data, label):
		data = self.scaler.inverse_transform(data)
		label = self.scaler.inverse_transform(label)
		
		_cnt = data.size
		_loss = nd.sum((data - label) ** 2).as_in_context(mx.cpu())
		if self.cnt is None:
			self.cnt = 0
			self.loss = 0
		self.cnt += _cnt
		self.loss += _loss