class MXNet::NDArray::Ops

Defined in:

Class Method Summary

Class Method Detail

def self._abs(data : MXNet::NDArray?, **kwargs) #

def self._Activation(data : MXNet::NDArray?, act_type, **kwargs) #

def self._adam_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, mean : MXNet::NDArray?, var : MXNet::NDArray?, lr, **kwargs) #

def self._add_n(args : Array(MXNet::NDArray), **kwargs) #

def self._arccos(data : MXNet::NDArray?, **kwargs) #

def self._arccosh(data : MXNet::NDArray?, **kwargs) #

def self._arcsin(data : MXNet::NDArray?, **kwargs) #

def self._arcsinh(data : MXNet::NDArray?, **kwargs) #

def self._arctan(data : MXNet::NDArray?, **kwargs) #

def self._arctanh(data : MXNet::NDArray?, **kwargs) #

def self._argmax(data : MXNet::NDArray?, **kwargs) #

def self._argmax_channel(data : MXNet::NDArray?, **kwargs) #

def self._argmin(data : MXNet::NDArray?, **kwargs) #

def self._argsort(data : MXNet::NDArray?, **kwargs) #

def self._batch_dot(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._batch_take(a : MXNet::NDArray?, indices : MXNet::NDArray?, **kwargs) #

def self._BatchNorm(data : MXNet::NDArray?, gamma : MXNet::NDArray?, beta : MXNet::NDArray?, moving_mean : MXNet::NDArray?, moving_var : MXNet::NDArray?, **kwargs) #

def self._BatchNorm_v1(data : MXNet::NDArray?, gamma : MXNet::NDArray?, beta : MXNet::NDArray?, **kwargs) #

def self._BilinearSampler(data : MXNet::NDArray?, grid : MXNet::NDArray?, **kwargs) #

def self._BlockGrad(data : MXNet::NDArray?, **kwargs) #

def self._broadcast_add(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_axes(data : MXNet::NDArray?, **kwargs) #

def self._broadcast_axis(data : MXNet::NDArray?, **kwargs) #

def self._broadcast_div(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_equal(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_greater(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_greater_equal(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_hypot(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_lesser(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_lesser_equal(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_like(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_logical_and(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_logical_or(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_logical_xor(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_maximum(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_minimum(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_minus(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_mod(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_mul(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_not_equal(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_plus(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_power(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_sub(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._broadcast_to(data : MXNet::NDArray?, **kwargs) #

def self._Cast(data : MXNet::NDArray?, dtype, **kwargs) #

def self._cast(data : MXNet::NDArray?, dtype, **kwargs) #

def self._cast_storage(data : MXNet::NDArray?, stype, **kwargs) #

def self._cbrt(data : MXNet::NDArray?, **kwargs) #

def self._ceil(data : MXNet::NDArray?, **kwargs) #

def self._choose_element_0index(**kwargs) #

def self._clip(data : MXNet::NDArray?, a_min, a_max, **kwargs) #

def self._Concat(data : Array(MXNet::NDArray), num_args, **kwargs) #

def self._concat(data : Array(MXNet::NDArray), num_args, **kwargs) #

def self._Convolution(data : MXNet::NDArray?, weight : MXNet::NDArray?, bias : MXNet::NDArray?, kernel, num_filter, **kwargs) #

def self._Convolution_v1(data : MXNet::NDArray?, weight : MXNet::NDArray?, bias : MXNet::NDArray?, kernel, num_filter, **kwargs) #

def self._Correlation(data1 : MXNet::NDArray?, data2 : MXNet::NDArray?, **kwargs) #

def self._cos(data : MXNet::NDArray?, **kwargs) #

def self._cosh(data : MXNet::NDArray?, **kwargs) #

def self._Crop(num_args, **kwargs) #

def self._crop(data : MXNet::NDArray?, begin _begin, end _end, **kwargs) #

def self._CuDNNBatchNorm(data : MXNet::NDArray?, gamma : MXNet::NDArray?, beta : MXNet::NDArray?, moving_mean : MXNet::NDArray?, moving_var : MXNet::NDArray?, **kwargs) #

def self._Custom(data : Array(MXNet::NDArray), op_type, **kwargs) #

def self._Deconvolution(data : MXNet::NDArray?, weight : MXNet::NDArray?, bias : MXNet::NDArray?, kernel, num_filter, **kwargs) #

def self._degrees(data : MXNet::NDArray?, **kwargs) #

def self._depth_to_space(data : MXNet::NDArray?, block_size, **kwargs) #

def self._diag(data : MXNet::NDArray?, **kwargs) #

def self._dot(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._Dropout(data : MXNet::NDArray?, **kwargs) #

def self._ElementWiseSum(args : Array(MXNet::NDArray), **kwargs) #

def self._elemwise_add(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._elemwise_div(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._elemwise_mul(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._elemwise_sub(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._Embedding(data : MXNet::NDArray?, weight : MXNet::NDArray?, input_dim, output_dim, **kwargs) #

def self._exp(data : MXNet::NDArray?, **kwargs) #

def self._expand_dims(data : MXNet::NDArray?, axis, **kwargs) #

def self._expm1(data : MXNet::NDArray?, **kwargs) #

def self._fill_element_0index(**kwargs) #

def self._fix(data : MXNet::NDArray?, **kwargs) #

def self._Flatten(data : MXNet::NDArray?, **kwargs) #

def self._flatten(data : MXNet::NDArray?, **kwargs) #

def self._flip(data : MXNet::NDArray?, axis, **kwargs) #

def self._floor(data : MXNet::NDArray?, **kwargs) #

def self._ftml_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, d : MXNet::NDArray?, v : MXNet::NDArray?, z : MXNet::NDArray?, lr, t, **kwargs) #

def self._ftrl_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, z : MXNet::NDArray?, n : MXNet::NDArray?, lr, **kwargs) #

def self._FullyConnected(data : MXNet::NDArray?, weight : MXNet::NDArray?, bias : MXNet::NDArray?, num_hidden, **kwargs) #

def self._gamma(data : MXNet::NDArray?, **kwargs) #

def self._gammaln(data : MXNet::NDArray?, **kwargs) #

def self._gather_nd(data : MXNet::NDArray?, indices : MXNet::NDArray?, **kwargs) #

def self._GridGenerator(data : MXNet::NDArray?, transform_type, **kwargs) #

def self._hard_sigmoid(data : MXNet::NDArray?, **kwargs) #

def self._identity(data : MXNet::NDArray?, **kwargs) #

def self._IdentityAttachKLSparseReg(data : MXNet::NDArray?, **kwargs) #

def self._InstanceNorm(data : MXNet::NDArray?, gamma : MXNet::NDArray?, beta : MXNet::NDArray?, **kwargs) #

def self._khatri_rao(args : Array(MXNet::NDArray), **kwargs) #

def self._L2Normalization(data : MXNet::NDArray?, **kwargs) #

def self._LayerNorm(data : MXNet::NDArray?, gamma : MXNet::NDArray?, beta : MXNet::NDArray?, **kwargs) #

def self._LeakyReLU(data : MXNet::NDArray?, gamma : MXNet::NDArray?, **kwargs) #

def self._linalg_gelqf(a : MXNet::NDArray?, **kwargs) #

def self._linalg_gemm(a : MXNet::NDArray?, b : MXNet::NDArray?, c : MXNet::NDArray?, **kwargs) #

def self._linalg_gemm2(a : MXNet::NDArray?, b : MXNet::NDArray?, **kwargs) #

def self._linalg_potrf(a : MXNet::NDArray?, **kwargs) #

def self._linalg_potri(a : MXNet::NDArray?, **kwargs) #

def self._linalg_sumlogdiag(a : MXNet::NDArray?, **kwargs) #

def self._linalg_syrk(a : MXNet::NDArray?, **kwargs) #

def self._linalg_trmm(a : MXNet::NDArray?, b : MXNet::NDArray?, **kwargs) #

def self._linalg_trsm(a : MXNet::NDArray?, b : MXNet::NDArray?, **kwargs) #

def self._LinearRegressionOutput(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._log(data : MXNet::NDArray?, **kwargs) #

def self._log10(data : MXNet::NDArray?, **kwargs) #

def self._log1p(data : MXNet::NDArray?, **kwargs) #

def self._log2(data : MXNet::NDArray?, **kwargs) #

def self._log_softmax(data : MXNet::NDArray?, **kwargs) #

def self._logical_not(data : MXNet::NDArray?, **kwargs) #

def self._LogisticRegressionOutput(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._LRN(data : MXNet::NDArray?, nsize, **kwargs) #

def self._MAERegressionOutput(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._make_loss(data : MXNet::NDArray?, **kwargs) #

def self._MakeLoss(data : MXNet::NDArray?, **kwargs) #

def self._max(data : MXNet::NDArray?, **kwargs) #

def self._max_axis(data : MXNet::NDArray?, **kwargs) #

def self._mean(data : MXNet::NDArray?, **kwargs) #

def self._min(data : MXNet::NDArray?, **kwargs) #

def self._min_axis(data : MXNet::NDArray?, **kwargs) #

def self._mp_sgd_mom_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, mom : MXNet::NDArray?, weight32 : MXNet::NDArray?, lr, **kwargs) #

def self._mp_sgd_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, weight32 : MXNet::NDArray?, lr, **kwargs) #

def self._nanprod(data : MXNet::NDArray?, **kwargs) #

def self._nansum(data : MXNet::NDArray?, **kwargs) #

def self._negative(data : MXNet::NDArray?, **kwargs) #

def self._norm(data : MXNet::NDArray?, **kwargs) #

def self._normal(**kwargs) #

def self._one_hot(indices : MXNet::NDArray?, depth, **kwargs) #

def self._ones_like(data : MXNet::NDArray?, **kwargs) #

def self._Pad(data : MXNet::NDArray?, mode, pad_width, **kwargs) #

def self._pad(data : MXNet::NDArray?, mode, pad_width, **kwargs) #

def self._pick(data : MXNet::NDArray?, index : MXNet::NDArray?, **kwargs) #

def self._Pooling(data : MXNet::NDArray?, **kwargs) #

def self._Pooling_v1(data : MXNet::NDArray?, **kwargs) #

def self._prod(data : MXNet::NDArray?, **kwargs) #

def self._radians(data : MXNet::NDArray?, **kwargs) #

def self._random_exponential(**kwargs) #

def self._random_gamma(**kwargs) #

def self._random_generalized_negative_binomial(**kwargs) #

def self._random_negative_binomial(**kwargs) #

def self._random_normal(**kwargs) #

def self._random_poisson(**kwargs) #

def self._random_randint(low, high, **kwargs) #

def self._random_uniform(**kwargs) #

def self._ravel_multi_index(data : MXNet::NDArray?, **kwargs) #

def self._rcbrt(data : MXNet::NDArray?, **kwargs) #

def self._reciprocal(data : MXNet::NDArray?, **kwargs) #

def self._relu(data : MXNet::NDArray?, **kwargs) #

def self._repeat(data : MXNet::NDArray?, repeats, **kwargs) #

def self._Reshape(data : MXNet::NDArray?, **kwargs) #

def self._reshape(data : MXNet::NDArray?, shape, **kwargs) #

def self._reshape_like(lhs : MXNet::NDArray?, rhs : MXNet::NDArray?, **kwargs) #

def self._reverse(data : MXNet::NDArray?, axis, **kwargs) #

def self._rint(data : MXNet::NDArray?, **kwargs) #

def self._rmsprop_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, n : MXNet::NDArray?, lr, **kwargs) #

def self._rmspropalex_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, n : MXNet::NDArray?, g : MXNet::NDArray?, delta : MXNet::NDArray?, lr, **kwargs) #

def self._RNN(data : MXNet::NDArray?, parameters : MXNet::NDArray?, state : MXNet::NDArray?, state_cell : MXNet::NDArray?, state_size, num_layers, mode, **kwargs) #

def self._ROIPooling(data : MXNet::NDArray?, rois : MXNet::NDArray?, pooled_size, spatial_scale, **kwargs) #

def self._round(data : MXNet::NDArray?, **kwargs) #

def self._rsqrt(data : MXNet::NDArray?, **kwargs) #

def self._sample_exponential(lam : MXNet::NDArray?, **kwargs) #

def self._sample_gamma(alpha : MXNet::NDArray?, beta : MXNet::NDArray?, **kwargs) #

def self._sample_generalized_negative_binomial(mu : MXNet::NDArray?, alpha : MXNet::NDArray?, **kwargs) #

def self._sample_multinomial(data : MXNet::NDArray?, **kwargs) #

def self._sample_negative_binomial(k : MXNet::NDArray?, p : MXNet::NDArray?, **kwargs) #

def self._sample_normal(mu : MXNet::NDArray?, sigma : MXNet::NDArray?, **kwargs) #

def self._sample_poisson(lam : MXNet::NDArray?, **kwargs) #

def self._sample_uniform(low : MXNet::NDArray?, high : MXNet::NDArray?, **kwargs) #

def self._scatter_nd(data : MXNet::NDArray?, indices : MXNet::NDArray?, shape, **kwargs) #

def self._SequenceLast(data : MXNet::NDArray?, sequence_length : MXNet::NDArray?, **kwargs) #

def self._SequenceMask(data : MXNet::NDArray?, sequence_length : MXNet::NDArray?, **kwargs) #

def self._SequenceReverse(data : MXNet::NDArray?, sequence_length : MXNet::NDArray?, **kwargs) #

def self._sgd_mom_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, mom : MXNet::NDArray?, lr, **kwargs) #

def self._sgd_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, lr, **kwargs) #

def self._shape_array(data : MXNet::NDArray?, **kwargs) #

def self._shuffle(data : MXNet::NDArray?, **kwargs) #

def self._sigmoid(data : MXNet::NDArray?, **kwargs) #

def self._sign(data : MXNet::NDArray?, **kwargs) #

def self._signsgd_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, lr, **kwargs) #

def self._signum_update(weight : MXNet::NDArray?, grad : MXNet::NDArray?, mom : MXNet::NDArray?, lr, **kwargs) #

def self._sin(data : MXNet::NDArray?, **kwargs) #

def self._sinh(data : MXNet::NDArray?, **kwargs) #

def self._size_array(data : MXNet::NDArray?, **kwargs) #

def self._slice(data : MXNet::NDArray?, begin _begin, end _end, **kwargs) #

def self._slice_axis(data : MXNet::NDArray?, axis, begin _begin, end _end, **kwargs) #

def self._slice_like(data : MXNet::NDArray?, shape_like : MXNet::NDArray?, **kwargs) #

def self._SliceChannel(data : MXNet::NDArray?, num_outputs, **kwargs) #

def self._smooth_l1(data : MXNet::NDArray?, scalar, **kwargs) #

def self._Softmax(data : MXNet::NDArray?, **kwargs) #

def self._softmax(data : MXNet::NDArray?, **kwargs) #

def self._softmax_cross_entropy(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._SoftmaxActivation(data : MXNet::NDArray?, **kwargs) #

def self._SoftmaxOutput(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._softsign(data : MXNet::NDArray?, **kwargs) #

def self._sort(data : MXNet::NDArray?, **kwargs) #

def self._space_to_depth(data : MXNet::NDArray?, block_size, **kwargs) #

def self._SpatialTransformer(data : MXNet::NDArray?, loc : MXNet::NDArray?, transform_type, sampler_type, **kwargs) #

def self._split(data : MXNet::NDArray?, num_outputs, **kwargs) #

def self._sqrt(data : MXNet::NDArray?, **kwargs) #

def self._square(data : MXNet::NDArray?, **kwargs) #

def self._squeeze(data : Array(MXNet::NDArray), **kwargs) #

def self._stack(data : Array(MXNet::NDArray), num_args, **kwargs) #

def self._stop_gradient(data : MXNet::NDArray?, **kwargs) #

def self._sum(data : MXNet::NDArray?, **kwargs) #

def self._sum_axis(data : MXNet::NDArray?, **kwargs) #

def self._SVMOutput(data : MXNet::NDArray?, label : MXNet::NDArray?, **kwargs) #

def self._swapaxes(data : MXNet::NDArray?, **kwargs) #

def self._SwapAxis(data : MXNet::NDArray?, **kwargs) #

def self._take(a : MXNet::NDArray?, indices : MXNet::NDArray?, **kwargs) #

def self._tan(data : MXNet::NDArray?, **kwargs) #

def self._tanh(data : MXNet::NDArray?, **kwargs) #

def self._tile(data : MXNet::NDArray?, reps, **kwargs) #

def self._topk(data : MXNet::NDArray?, **kwargs) #

def self._transpose(data : MXNet::NDArray?, **kwargs) #

def self._trunc(data : MXNet::NDArray?, **kwargs) #

def self._uniform(**kwargs) #

def self._unravel_index(data : MXNet::NDArray?, **kwargs) #

def self._UpSampling(data : Array(MXNet::NDArray), scale, sample_type, num_args, **kwargs) #

def self._where(condition : MXNet::NDArray?, x : MXNet::NDArray?, y : MXNet::NDArray?, **kwargs) #

def self._zeros_like(data : MXNet::NDArray?, **kwargs) #