MXNet
-
- ndarray
- ndarray.CachedOp
- ndarray.NDArray
- ndarray.Activation
- ndarray.BatchNorm
- ndarray.BatchNorm_v1
- ndarray.BilinearSampler
- ndarray.BlockGrad
- ndarray.CTCLoss
- ndarray.Cast
- ndarray.Concat
- ndarray.Convolution
- ndarray.Convolution_v1
- ndarray.Correlation
- ndarray.Crop
- ndarray.Custom
- ndarray.Deconvolution
- ndarray.Dropout
- ndarray.ElementWiseSum
- ndarray.Embedding
- ndarray.Flatten
- ndarray.FullyConnected
- ndarray.GridGenerator
- ndarray.GroupNorm
- ndarray.IdentityAttachKLSparseReg
- ndarray.InstanceNorm
- ndarray.L2Normalization
- ndarray.LRN
- ndarray.LayerNorm
- ndarray.LeakyReLU
- ndarray.LinearRegressionOutput
- ndarray.LogisticRegressionOutput
- ndarray.MAERegressionOutput
- ndarray.MakeLoss
- ndarray.Pad
- ndarray.Pooling
- ndarray.Pooling_v1
- ndarray.RNN
- ndarray.ROIPooling
- ndarray.Reshape
- ndarray.SVMOutput
- ndarray.SequenceLast
- ndarray.SequenceMask
- ndarray.SequenceReverse
- ndarray.SliceChannel
- ndarray.Softmax
- ndarray.SoftmaxActivation
- ndarray.SoftmaxOutput
- ndarray.SpatialTransformer
- ndarray.SwapAxis
- ndarray.UpSampling
- ndarray.abs
- ndarray.adam_update
- ndarray.add_n
- ndarray.all_finite
- ndarray.amp_cast
- ndarray.amp_multicast
- ndarray.arccos
- ndarray.arccosh
- ndarray.arcsin
- ndarray.arcsinh
- ndarray.arctan
- ndarray.arctanh
- ndarray.argmax
- ndarray.argmax_channel
- ndarray.argmin
- ndarray.argsort
- ndarray.batch_dot
- ndarray.batch_take
- ndarray.broadcast_add
- ndarray.broadcast_axes
- ndarray.broadcast_axis
- ndarray.broadcast_div
- ndarray.broadcast_equal
- ndarray.broadcast_greater
- ndarray.broadcast_greater_equal
- ndarray.broadcast_hypot
- ndarray.broadcast_lesser
- ndarray.broadcast_lesser_equal
- ndarray.broadcast_like
- ndarray.broadcast_logical_and
- ndarray.broadcast_logical_or
- ndarray.broadcast_logical_xor
- ndarray.broadcast_maximum
- ndarray.broadcast_minimum
- ndarray.broadcast_minus
- ndarray.broadcast_mod
- ndarray.broadcast_mul
- ndarray.broadcast_not_equal
- ndarray.broadcast_plus
- ndarray.broadcast_power
- ndarray.broadcast_sub
- ndarray.broadcast_to
- ndarray.cast
- ndarray.cast_storage
- ndarray.cbrt
- ndarray.ceil
- ndarray.choose_element_0index
- ndarray.clip
- ndarray.col2im
- ndarray.concat
- ndarray.cos
- ndarray.cosh
- ndarray.crop
- ndarray.ctc_loss
- ndarray.cumsum
- ndarray.degrees
- ndarray.depth_to_space
- ndarray.diag
- ndarray.dot
- ndarray.elemwise_add
- ndarray.elemwise_div
- ndarray.elemwise_mul
- ndarray.elemwise_sub
- ndarray.erf
- ndarray.erfinv
- ndarray.exp
- ndarray.expand_dims
- ndarray.expm1
- ndarray.fill_element_0index
- ndarray.fix
- ndarray.flatten
- ndarray.flip
- ndarray.floor
- ndarray.ftml_update
- ndarray.ftrl_update
- ndarray.gamma
- ndarray.gammaln
- ndarray.gather_nd
- ndarray.hard_sigmoid
- ndarray.identity
- ndarray.im2col
- ndarray.khatri_rao
- ndarray.lamb_update_phase1
- ndarray.lamb_update_phase2
- ndarray.linalg_det
- ndarray.linalg_extractdiag
- ndarray.linalg_extracttrian
- ndarray.linalg_gelqf
- ndarray.linalg_gemm
- ndarray.linalg_gemm2
- ndarray.linalg_inverse
- ndarray.linalg_makediag
- ndarray.linalg_maketrian
- ndarray.linalg_potrf
- ndarray.linalg_potri
- ndarray.linalg_slogdet
- ndarray.linalg_sumlogdiag
- ndarray.linalg_syrk
- ndarray.linalg_trmm
- ndarray.linalg_trsm
- ndarray.log
- ndarray.log10
- ndarray.log1p
- ndarray.log2
- ndarray.log_softmax
- ndarray.logical_not
- ndarray.make_loss
- ndarray.max
- ndarray.max_axis
- ndarray.mean
- ndarray.min
- ndarray.min_axis
- ndarray.moments
- ndarray.mp_lamb_update_phase1
- ndarray.mp_lamb_update_phase2
- ndarray.mp_nag_mom_update
- ndarray.mp_sgd_mom_update
- ndarray.mp_sgd_update
- ndarray.multi_all_finite
- ndarray.multi_lars
- ndarray.multi_mp_sgd_mom_update
- ndarray.multi_mp_sgd_update
- ndarray.multi_sgd_mom_update
- ndarray.multi_sgd_update
- ndarray.multi_sum_sq
- ndarray.nag_mom_update
- ndarray.nanprod
- ndarray.nansum
- ndarray.negative
- ndarray.norm
- ndarray.normal
- ndarray.one_hot
- ndarray.ones_like
- ndarray.pad
- ndarray.pick
- ndarray.preloaded_multi_mp_sgd_mom_update
- ndarray.preloaded_multi_mp_sgd_update
- ndarray.preloaded_multi_sgd_mom_update
- ndarray.preloaded_multi_sgd_update
- ndarray.prod
- ndarray.radians
- ndarray.random_exponential
- ndarray.random_gamma
- ndarray.random_generalized_negative_binomial
- ndarray.random_negative_binomial
- ndarray.random_normal
- ndarray.random_pdf_dirichlet
- ndarray.random_pdf_exponential
- ndarray.random_pdf_gamma
- ndarray.random_pdf_generalized_negative_binomial
- ndarray.random_pdf_negative_binomial
- ndarray.random_pdf_normal
- ndarray.random_pdf_poisson
- ndarray.random_pdf_uniform
- ndarray.random_poisson
- ndarray.random_randint
- ndarray.random_uniform
- ndarray.ravel_multi_index
- ndarray.rcbrt
- ndarray.reciprocal
- ndarray.relu
- ndarray.repeat
- ndarray.reset_arrays
- ndarray.reshape
- ndarray.reshape_like
- ndarray.reverse
- ndarray.rint
- ndarray.rmsprop_update
- ndarray.rmspropalex_update
- ndarray.round
- ndarray.rsqrt
- ndarray.sample_exponential
- ndarray.sample_gamma
- ndarray.sample_generalized_negative_binomial
- ndarray.sample_multinomial
- ndarray.sample_negative_binomial
- ndarray.sample_normal
- ndarray.sample_poisson
- ndarray.sample_uniform
- ndarray.scatter_nd
- ndarray.sgd_mom_update
- ndarray.sgd_update
- ndarray.shape_array
- ndarray.shuffle
- ndarray.sigmoid
- ndarray.sign
- ndarray.signsgd_update
- ndarray.signum_update
- ndarray.sin
- ndarray.sinh
- ndarray.size_array
- ndarray.slice
- ndarray.slice_axis
- ndarray.slice_like
- ndarray.smooth_l1
- ndarray.softmax
- ndarray.softmax_cross_entropy
- ndarray.softmin
- ndarray.softsign
- ndarray.sort
- ndarray.space_to_depth
- ndarray.split
- ndarray.sqrt
- ndarray.square
- ndarray.squeeze
- ndarray.stack
- ndarray.stop_gradient
- ndarray.sum
- ndarray.sum_axis
- ndarray.swapaxes
- ndarray.take
- ndarray.tan
- ndarray.tanh
- ndarray.tile
- ndarray.topk
- ndarray.transpose
- ndarray.trunc
- ndarray.uniform
- ndarray.unravel_index
- ndarray.where
- ndarray.zeros_like
- ndarray.concatenate
- ndarray.ones
- ndarray.add
- ndarray.arange
- ndarray.linspace
- ndarray.eye
- ndarray.divide
- ndarray.equal
- ndarray.full
- ndarray.greater
- ndarray.greater_equal
- ndarray.imdecode
- ndarray.lesser
- ndarray.lesser_equal
- ndarray.logical_and
- ndarray.logical_or
- ndarray.logical_xor
- ndarray.maximum
- ndarray.minimum
- ndarray.moveaxis
- ndarray.modulo
- ndarray.multiply
- ndarray.not_equal
- ndarray.onehot_encode
- ndarray.power
- ndarray.subtract
- ndarray.true_divide
- ndarray.waitall
- ndarray.histogram
- ndarray.split_v2
- ndarray.to_dlpack_for_read
- ndarray.to_dlpack_for_write
- ndarray.from_dlpack
- ndarray.from_numpy
- ndarray.zeros
- ndarray.indexing_key_expand_implicit_axes
- ndarray.get_indexing_dispatch_code
- ndarray.get_oshape_of_gather_nd_op
- ndarray.empty
- ndarray.array
- ndarray.load
- ndarray.load_frombuffer
- ndarray.save
-
- ndarray.contrib
- ndarray.contrib.rand_zipfian
- ndarray.contrib.foreach
- ndarray.contrib.while_loop
- ndarray.contrib.cond
- ndarray.contrib.isinf
- ndarray.contrib.isfinite
- ndarray.contrib.isnan
- ndarray.contrib.AdaptiveAvgPooling2D
- ndarray.contrib.BilinearResize2D
- ndarray.contrib.CTCLoss
- ndarray.contrib.DeformableConvolution
- ndarray.contrib.DeformablePSROIPooling
- ndarray.contrib.ModulatedDeformableConvolution
- ndarray.contrib.MultiBoxDetection
- ndarray.contrib.MultiBoxPrior
- ndarray.contrib.MultiBoxTarget
- ndarray.contrib.MultiProposal
- ndarray.contrib.PSROIPooling
- ndarray.contrib.Proposal
- ndarray.contrib.ROIAlign
- ndarray.contrib.RROIAlign
- ndarray.contrib.SparseEmbedding
- ndarray.contrib.SyncBatchNorm
- ndarray.contrib.allclose
- ndarray.contrib.arange_like
- ndarray.contrib.backward_gradientmultiplier
- ndarray.contrib.backward_hawkesll
- ndarray.contrib.backward_index_copy
- ndarray.contrib.backward_quadratic
- ndarray.contrib.bipartite_matching
- ndarray.contrib.boolean_mask
- ndarray.contrib.box_decode
- ndarray.contrib.box_encode
- ndarray.contrib.box_iou
- ndarray.contrib.box_nms
- ndarray.contrib.box_non_maximum_suppression
- ndarray.contrib.calibrate_entropy
- ndarray.contrib.count_sketch
- ndarray.contrib.ctc_loss
- ndarray.contrib.dequantize
- ndarray.contrib.dgl_adjacency
- ndarray.contrib.dgl_csr_neighbor_non_uniform_sample
- ndarray.contrib.dgl_csr_neighbor_uniform_sample
- ndarray.contrib.dgl_graph_compact
- ndarray.contrib.dgl_subgraph
- ndarray.contrib.div_sqrt_dim
- ndarray.contrib.edge_id
- ndarray.contrib.fft
- ndarray.contrib.getnnz
- ndarray.contrib.gradientmultiplier
- ndarray.contrib.group_adagrad_update
- ndarray.contrib.hawkesll
- ndarray.contrib.ifft
- ndarray.contrib.index_array
- ndarray.contrib.index_copy
- ndarray.contrib.interleaved_matmul_encdec_qk
- ndarray.contrib.interleaved_matmul_encdec_valatt
- ndarray.contrib.interleaved_matmul_selfatt_qk
- ndarray.contrib.interleaved_matmul_selfatt_valatt
- ndarray.contrib.quadratic
- ndarray.contrib.quantize
- ndarray.contrib.quantize_v2
- ndarray.contrib.quantized_act
- ndarray.contrib.quantized_batch_norm
- ndarray.contrib.quantized_concat
- ndarray.contrib.quantized_conv
- ndarray.contrib.quantized_elemwise_add
- ndarray.contrib.quantized_elemwise_mul
- ndarray.contrib.quantized_embedding
- ndarray.contrib.quantized_flatten
- ndarray.contrib.quantized_fully_connected
- ndarray.contrib.quantized_pooling
- ndarray.contrib.requantize
- ndarray.contrib.round_ste
- ndarray.contrib.sign_ste
-
- ndarray.image
- ndarray.image.adjust_lighting
- ndarray.image.crop
- ndarray.image.flip_left_right
- ndarray.image.flip_top_bottom
- ndarray.image.normalize
- ndarray.image.random_brightness
- ndarray.image.random_color_jitter
- ndarray.image.random_contrast
- ndarray.image.random_flip_left_right
- ndarray.image.random_flip_top_bottom
- ndarray.image.random_hue
- ndarray.image.random_lighting
- ndarray.image.random_saturation
- ndarray.image.resize
- ndarray.image.to_tensor
-
- ndarray.linalg
- ndarray.linalg.det
- ndarray.linalg.extractdiag
- ndarray.linalg.extracttrian
- ndarray.linalg.gelqf
- ndarray.linalg.gemm
- ndarray.linalg.gemm2
- ndarray.linalg.inverse
- ndarray.linalg.makediag
- ndarray.linalg.maketrian
- ndarray.linalg.potrf
- ndarray.linalg.potri
- ndarray.linalg.slogdet
- ndarray.linalg.sumlogdiag
- ndarray.linalg.syevd
- ndarray.linalg.syrk
- ndarray.linalg.trmm
- ndarray.linalg.trsm
-
- ndarray.op
- ndarray.op.CachedOp
- ndarray.op.Activation
- ndarray.op.BatchNorm
- ndarray.op.BatchNorm_v1
- ndarray.op.BilinearSampler
- ndarray.op.BlockGrad
- ndarray.op.CTCLoss
- ndarray.op.Cast
- ndarray.op.Concat
- ndarray.op.Convolution
- ndarray.op.Convolution_v1
- ndarray.op.Correlation
- ndarray.op.Crop
- ndarray.op.Custom
- ndarray.op.Deconvolution
- ndarray.op.Dropout
- ndarray.op.ElementWiseSum
- ndarray.op.Embedding
- ndarray.op.Flatten
- ndarray.op.FullyConnected
- ndarray.op.GridGenerator
- ndarray.op.GroupNorm
- ndarray.op.IdentityAttachKLSparseReg
- ndarray.op.InstanceNorm
- ndarray.op.L2Normalization
- ndarray.op.LRN
- ndarray.op.LayerNorm
- ndarray.op.LeakyReLU
- ndarray.op.LinearRegressionOutput
- ndarray.op.LogisticRegressionOutput
- ndarray.op.MAERegressionOutput
- ndarray.op.MakeLoss
- ndarray.op.Pad
- ndarray.op.Pooling
- ndarray.op.Pooling_v1
- ndarray.op.RNN
- ndarray.op.ROIPooling
- ndarray.op.Reshape
- ndarray.op.SVMOutput
- ndarray.op.SequenceLast
- ndarray.op.SequenceMask
- ndarray.op.SequenceReverse
- ndarray.op.SliceChannel
- ndarray.op.Softmax
- ndarray.op.SoftmaxActivation
- ndarray.op.SoftmaxOutput
- ndarray.op.SpatialTransformer
- ndarray.op.SwapAxis
- ndarray.op.UpSampling
- ndarray.op.abs
- ndarray.op.adam_update
- ndarray.op.add_n
- ndarray.op.all_finite
- ndarray.op.amp_cast
- ndarray.op.amp_multicast
- ndarray.op.arccos
- ndarray.op.arccosh
- ndarray.op.arcsin
- ndarray.op.arcsinh
- ndarray.op.arctan
- ndarray.op.arctanh
- ndarray.op.argmax
- ndarray.op.argmax_channel
- ndarray.op.argmin
- ndarray.op.argsort
- ndarray.op.batch_dot
- ndarray.op.batch_take
- ndarray.op.broadcast_add
- ndarray.op.broadcast_axes
- ndarray.op.broadcast_axis
- ndarray.op.broadcast_div
- ndarray.op.broadcast_equal
- ndarray.op.broadcast_greater
- ndarray.op.broadcast_greater_equal
- ndarray.op.broadcast_hypot
- ndarray.op.broadcast_lesser
- ndarray.op.broadcast_lesser_equal
- ndarray.op.broadcast_like
- ndarray.op.broadcast_logical_and
- ndarray.op.broadcast_logical_or
- ndarray.op.broadcast_logical_xor
- ndarray.op.broadcast_maximum
- ndarray.op.broadcast_minimum
- ndarray.op.broadcast_minus
- ndarray.op.broadcast_mod
- ndarray.op.broadcast_mul
- ndarray.op.broadcast_not_equal
- ndarray.op.broadcast_plus
- ndarray.op.broadcast_power
- ndarray.op.broadcast_sub
- ndarray.op.broadcast_to
- ndarray.op.cast
- ndarray.op.cast_storage
- ndarray.op.cbrt
- ndarray.op.ceil
- ndarray.op.choose_element_0index
- ndarray.op.clip
- ndarray.op.col2im
- ndarray.op.concat
- ndarray.op.cos
- ndarray.op.cosh
- ndarray.op.crop
- ndarray.op.ctc_loss
- ndarray.op.cumsum
- ndarray.op.degrees
- ndarray.op.depth_to_space
- ndarray.op.diag
- ndarray.op.dot
- ndarray.op.elemwise_add
- ndarray.op.elemwise_div
- ndarray.op.elemwise_mul
- ndarray.op.elemwise_sub
- ndarray.op.erf
- ndarray.op.erfinv
- ndarray.op.exp
- ndarray.op.expand_dims
- ndarray.op.expm1
- ndarray.op.fill_element_0index
- ndarray.op.fix
- ndarray.op.flatten
- ndarray.op.flip
- ndarray.op.floor
- ndarray.op.ftml_update
- ndarray.op.ftrl_update
- ndarray.op.gamma
- ndarray.op.gammaln
- ndarray.op.gather_nd
- ndarray.op.hard_sigmoid
- ndarray.op.identity
- ndarray.op.im2col
- ndarray.op.khatri_rao
- ndarray.op.lamb_update_phase1
- ndarray.op.lamb_update_phase2
- ndarray.op.linalg_det
- ndarray.op.linalg_extractdiag
- ndarray.op.linalg_extracttrian
- ndarray.op.linalg_gelqf
- ndarray.op.linalg_gemm
- ndarray.op.linalg_gemm2
- ndarray.op.linalg_inverse
- ndarray.op.linalg_makediag
- ndarray.op.linalg_maketrian
- ndarray.op.linalg_potrf
- ndarray.op.linalg_potri
- ndarray.op.linalg_slogdet
- ndarray.op.linalg_sumlogdiag
- ndarray.op.linalg_syrk
- ndarray.op.linalg_trmm
- ndarray.op.linalg_trsm
- ndarray.op.log
- ndarray.op.log10
- ndarray.op.log1p
- ndarray.op.log2
- ndarray.op.log_softmax
- ndarray.op.logical_not
- ndarray.op.make_loss
- ndarray.op.max
- ndarray.op.max_axis
- ndarray.op.mean
- ndarray.op.min
- ndarray.op.min_axis
- ndarray.op.moments
- ndarray.op.mp_lamb_update_phase1
- ndarray.op.mp_lamb_update_phase2
- ndarray.op.mp_nag_mom_update
- ndarray.op.mp_sgd_mom_update
- ndarray.op.mp_sgd_update
- ndarray.op.multi_all_finite
- ndarray.op.multi_lars
- ndarray.op.multi_mp_sgd_mom_update
- ndarray.op.multi_mp_sgd_update
- ndarray.op.multi_sgd_mom_update
- ndarray.op.multi_sgd_update
- ndarray.op.multi_sum_sq
- ndarray.op.nag_mom_update
- ndarray.op.nanprod
- ndarray.op.nansum
- ndarray.op.negative
- ndarray.op.norm
- ndarray.op.normal
- ndarray.op.one_hot
- ndarray.op.ones_like
- ndarray.op.pad
- ndarray.op.pick
- ndarray.op.preloaded_multi_mp_sgd_mom_update
- ndarray.op.preloaded_multi_mp_sgd_update
- ndarray.op.preloaded_multi_sgd_mom_update
- ndarray.op.preloaded_multi_sgd_update
- ndarray.op.prod
- ndarray.op.radians
- ndarray.op.random_exponential
- ndarray.op.random_gamma
- ndarray.op.random_generalized_negative_binomial
- ndarray.op.random_negative_binomial
- ndarray.op.random_normal
- ndarray.op.random_pdf_dirichlet
- ndarray.op.random_pdf_exponential
- ndarray.op.random_pdf_gamma
- ndarray.op.random_pdf_generalized_negative_binomial
- ndarray.op.random_pdf_negative_binomial
- ndarray.op.random_pdf_normal
- ndarray.op.random_pdf_poisson
- ndarray.op.random_pdf_uniform
- ndarray.op.random_poisson
- ndarray.op.random_randint
- ndarray.op.random_uniform
- ndarray.op.ravel_multi_index
- ndarray.op.rcbrt
- ndarray.op.reciprocal
- ndarray.op.relu
- ndarray.op.repeat
- ndarray.op.reset_arrays
- ndarray.op.reshape
- ndarray.op.reshape_like
- ndarray.op.reverse
- ndarray.op.rint
- ndarray.op.rmsprop_update
- ndarray.op.rmspropalex_update
- ndarray.op.round
- ndarray.op.rsqrt
- ndarray.op.sample_exponential
- ndarray.op.sample_gamma
- ndarray.op.sample_generalized_negative_binomial
- ndarray.op.sample_multinomial
- ndarray.op.sample_negative_binomial
- ndarray.op.sample_normal
- ndarray.op.sample_poisson
- ndarray.op.sample_uniform
- ndarray.op.scatter_nd
- ndarray.op.sgd_mom_update
- ndarray.op.sgd_update
- ndarray.op.shape_array
- ndarray.op.shuffle
- ndarray.op.sigmoid
- ndarray.op.sign
- ndarray.op.signsgd_update
- ndarray.op.signum_update
- ndarray.op.sin
- ndarray.op.sinh
- ndarray.op.size_array
- ndarray.op.slice
- ndarray.op.slice_axis
- ndarray.op.slice_like
- ndarray.op.smooth_l1
- ndarray.op.softmax
- ndarray.op.softmax_cross_entropy
- ndarray.op.softmin
- ndarray.op.softsign
- ndarray.op.sort
- ndarray.op.space_to_depth
- ndarray.op.split
- ndarray.op.sqrt
- ndarray.op.square
- ndarray.op.squeeze
- ndarray.op.stack
- ndarray.op.stop_gradient
- ndarray.op.sum
- ndarray.op.sum_axis
- ndarray.op.swapaxes
- ndarray.op.take
- ndarray.op.tan
- ndarray.op.tanh
- ndarray.op.tile
- ndarray.op.topk
- ndarray.op.transpose
- ndarray.op.trunc
- ndarray.op.uniform
- ndarray.op.unravel_index
- ndarray.op.where
- ndarray.op.zeros_like
-
- ndarray.random
- ndarray.random.uniform
- ndarray.random.normal
- ndarray.random.randn
- ndarray.random.poisson
- ndarray.random.exponential
- ndarray.random.gamma
- ndarray.random.multinomial
- ndarray.random.negative_binomial
- ndarray.random.generalized_negative_binomial
- ndarray.random.shuffle
- ndarray.random.randint
- ndarray.random.exponential_like
- ndarray.random.gamma_like
- ndarray.random.generalized_negative_binomial_like
- ndarray.random.negative_binomial_like
- ndarray.random.normal_like
- ndarray.random.poisson_like
- ndarray.random.uniform_like
- ndarray.register
-
- ndarray.sparse
- ndarray.sparse.csr_matrix
- ndarray.sparse.row_sparse_array
- ndarray.sparse.add
- ndarray.sparse.subtract
- ndarray.sparse.multiply
- ndarray.sparse.divide
- ndarray.sparse.ElementWiseSum
- ndarray.sparse.Embedding
- ndarray.sparse.FullyConnected
- ndarray.sparse.LinearRegressionOutput
- ndarray.sparse.LogisticRegressionOutput
- ndarray.sparse.MAERegressionOutput
- ndarray.sparse.abs
- ndarray.sparse.adagrad_update
- ndarray.sparse.adam_update
- ndarray.sparse.add_n
- ndarray.sparse.arccos
- ndarray.sparse.arccosh
- ndarray.sparse.arcsin
- ndarray.sparse.arcsinh
- ndarray.sparse.arctan
- ndarray.sparse.arctanh
- ndarray.sparse.broadcast_add
- ndarray.sparse.broadcast_div
- ndarray.sparse.broadcast_minus
- ndarray.sparse.broadcast_mul
- ndarray.sparse.broadcast_plus
- ndarray.sparse.broadcast_sub
- ndarray.sparse.cast_storage
- ndarray.sparse.cbrt
- ndarray.sparse.ceil
- ndarray.sparse.clip
- ndarray.sparse.concat
- ndarray.sparse.cos
- ndarray.sparse.cosh
- ndarray.sparse.degrees
- ndarray.sparse.dot
- ndarray.sparse.elemwise_add
- ndarray.sparse.elemwise_div
- ndarray.sparse.elemwise_mul
- ndarray.sparse.elemwise_sub
- ndarray.sparse.exp
- ndarray.sparse.expm1
- ndarray.sparse.fix
- ndarray.sparse.floor
- ndarray.sparse.ftrl_update
- ndarray.sparse.gamma
- ndarray.sparse.gammaln
- ndarray.sparse.log
- ndarray.sparse.log10
- ndarray.sparse.log1p
- ndarray.sparse.log2
- ndarray.sparse.make_loss
- ndarray.sparse.mean
- ndarray.sparse.negative
- ndarray.sparse.norm
- ndarray.sparse.radians
- ndarray.sparse.relu
- ndarray.sparse.retain
- ndarray.sparse.rint
- ndarray.sparse.round
- ndarray.sparse.rsqrt
- ndarray.sparse.sgd_mom_update
- ndarray.sparse.sgd_update
- ndarray.sparse.sigmoid
- ndarray.sparse.sign
- ndarray.sparse.sin
- ndarray.sparse.sinh
- ndarray.sparse.slice
- ndarray.sparse.sqrt
- ndarray.sparse.square
- ndarray.sparse.stop_gradient
- ndarray.sparse.sum
- ndarray.sparse.tan
- ndarray.sparse.tanh
- ndarray.sparse.trunc
- ndarray.sparse.where
- ndarray.sparse.zeros_like
- ndarray.sparse.BaseSparseNDArray
- ndarray.sparse.CSRNDArray
- ndarray.sparse.RowSparseNDArray
-
- gluon.Block
- gluon.Block.apply
- gluon.Block.cast
- gluon.Block.collect_params
- gluon.Block.forward
- gluon.Block.hybridize
- gluon.Block.initialize
- gluon.Block.load_parameters
- gluon.Block.load_params
- gluon.Block.name_scope
- gluon.Block.register_child
- gluon.Block.register_forward_hook
- gluon.Block.register_forward_pre_hook
- gluon.Block.register_op_hook
- gluon.Block.save_parameters
- gluon.Block.save_params
- gluon.Block.summary
-
- gluon.HybridBlock
- gluon.HybridBlock.apply
- gluon.HybridBlock.cast
- gluon.HybridBlock.collect_params
- gluon.HybridBlock.export
- gluon.HybridBlock.forward
- gluon.HybridBlock.hybrid_forward
- gluon.HybridBlock.hybridize
- gluon.HybridBlock.infer_shape
- gluon.HybridBlock.infer_type
- gluon.HybridBlock.initialize
- gluon.HybridBlock.load_parameters
- gluon.HybridBlock.load_params
- gluon.HybridBlock.name_scope
- gluon.HybridBlock.optimize_for
- gluon.HybridBlock.register_child
- gluon.HybridBlock.register_forward_hook
- gluon.HybridBlock.register_forward_pre_hook
- gluon.HybridBlock.register_op_hook
- gluon.HybridBlock.save_parameters
- gluon.HybridBlock.save_params
- gluon.HybridBlock.summary
-
- gluon.SymbolBlock
- gluon.SymbolBlock.apply
- gluon.SymbolBlock.cast
- gluon.SymbolBlock.collect_params
- gluon.SymbolBlock.export
- gluon.SymbolBlock.forward
- gluon.SymbolBlock.hybrid_forward
- gluon.SymbolBlock.hybridize
- gluon.SymbolBlock.imports
- gluon.SymbolBlock.infer_shape
- gluon.SymbolBlock.infer_type
- gluon.SymbolBlock.initialize
- gluon.SymbolBlock.load_parameters
- gluon.SymbolBlock.load_params
- gluon.SymbolBlock.name_scope
- gluon.SymbolBlock.optimize_for
- gluon.SymbolBlock.register_child
- gluon.SymbolBlock.register_forward_hook
- gluon.SymbolBlock.register_forward_pre_hook
- gluon.SymbolBlock.register_op_hook
- gluon.SymbolBlock.save_parameters
- gluon.SymbolBlock.save_params
- gluon.SymbolBlock.summary
-
- gluon.Constant
- gluon.Constant.cast
- gluon.Constant.data
- gluon.Constant.grad
- gluon.Constant.initialize
- gluon.Constant.list_ctx
- gluon.Constant.list_data
- gluon.Constant.list_grad
- gluon.Constant.list_row_sparse_data
- gluon.Constant.reset_ctx
- gluon.Constant.row_sparse_data
- gluon.Constant.set_data
- gluon.Constant.var
- gluon.Constant.zero_grad
-
- gluon.Parameter
- gluon.Parameter.cast
- gluon.Parameter.data
- gluon.Parameter.grad
- gluon.Parameter.initialize
- gluon.Parameter.list_ctx
- gluon.Parameter.list_data
- gluon.Parameter.list_grad
- gluon.Parameter.list_row_sparse_data
- gluon.Parameter.reset_ctx
- gluon.Parameter.row_sparse_data
- gluon.Parameter.set_data
- gluon.Parameter.var
- gluon.Parameter.zero_grad
-
- gluon.ParameterDict
- gluon.ParameterDict.get
- gluon.ParameterDict.get_constant
- gluon.ParameterDict.initialize
- gluon.ParameterDict.list_ctx
- gluon.ParameterDict.load
- gluon.ParameterDict.load_dict
- gluon.ParameterDict.reset_ctx
- gluon.ParameterDict.save
- gluon.ParameterDict.setattr
- gluon.ParameterDict.update
- gluon.ParameterDict.zero_grad
- gluon.contrib
-
- gluon.data
- gluon.data.vision.datasets
- gluon.data.vision.transforms
- gluon.data.Dataset
- gluon.data.ArrayDataset
- gluon.data.RecordFileDataset
- gluon.data.SimpleDataset
- gluon.data.BatchSampler
- gluon.data.DataLoader
- gluon.data.FilterSampler
- gluon.data.RandomSampler
- gluon.data.Sampler
- gluon.data.SequentialSampler
-
- gluon.loss
- gluon.loss.Loss
- gluon.loss.L2Loss
- gluon.loss.L1Loss
- gluon.loss.SigmoidBinaryCrossEntropyLoss
- gluon.loss.SigmoidBCELoss
- gluon.loss.SoftmaxCrossEntropyLoss
- gluon.loss.SoftmaxCELoss
- gluon.loss.KLDivLoss
- gluon.loss.CTCLoss
- gluon.loss.HuberLoss
- gluon.loss.HingeLoss
- gluon.loss.SquaredHingeLoss
- gluon.loss.LogisticLoss
- gluon.loss.TripletLoss
- gluon.loss.PoissonNLLLoss
- gluon.loss.CosineEmbeddingLoss
- gluon.loss.SDMLLoss
- gluon.nn
- gluon.rnn
- initializer
- initializer.Bilinear
- initializer.Constant
- initializer.FusedRNN
- initializer.InitDesc
- initializer.Initializer
- initializer.LSTMBias
- initializer.Load
- initializer.MSRAPrelu
- initializer.Mixed
- initializer.Normal
- initializer.One
- initializer.Orthogonal
- initializer.Uniform
- initializer.Xavier
- initializer.Zero
- optimizer
- optimizer.AdaDelta
- optimizer.AdaGrad
- optimizer.Adam
- optimizer.Adamax
- optimizer.DCASGD
- optimizer.FTML
- optimizer.Ftrl
- optimizer.LARS
- optimizer.LBSGD
- optimizer.NAG
- optimizer.Nadam
- optimizer.Optimizer
- optimizer.RMSProp
- optimizer.SGD
- optimizer.SGLD
- optimizer.Signum
- optimizer.LAMB
- optimizer.Test
- optimizer.Updater
- optimizer.ccSGD
- metric
- metric.Accuracy
- metric.Caffe
- metric.CompositeEvalMetric
- metric.CrossEntropy
- metric.CustomMetric
- metric.EvalMetric
- metric.F1
- metric.Loss
- metric.MAE
- metric.MCC
- metric.MSE
- metric.NegativeLogLikelihood
- metric.PCC
- metric.PearsonCorrelation
- metric.Perplexity
- metric.RMSE
- metric.TopKAccuracy
- metric.Torch
- symbol
-
- symbol.contrib
- symbol.contrib.rand_zipfian
- symbol.contrib.foreach
- symbol.contrib.while_loop
- symbol.contrib.cond
- symbol.contrib.AdaptiveAvgPooling2D
- symbol.contrib.BilinearResize2D
- symbol.contrib.CTCLoss
- symbol.contrib.DeformableConvolution
- symbol.contrib.DeformablePSROIPooling
- symbol.contrib.ModulatedDeformableConvolution
- symbol.contrib.MultiBoxDetection
- symbol.contrib.MultiBoxPrior
- symbol.contrib.MultiBoxTarget
- symbol.contrib.MultiProposal
- symbol.contrib.PSROIPooling
- symbol.contrib.Proposal
- symbol.contrib.ROIAlign
- symbol.contrib.RROIAlign
- symbol.contrib.SparseEmbedding
- symbol.contrib.SyncBatchNorm
- symbol.contrib.allclose
- symbol.contrib.arange_like
- symbol.contrib.backward_gradientmultiplier
- symbol.contrib.backward_hawkesll
- symbol.contrib.backward_index_copy
- symbol.contrib.backward_quadratic
- symbol.contrib.bipartite_matching
- symbol.contrib.boolean_mask
- symbol.contrib.box_decode
- symbol.contrib.box_encode
- symbol.contrib.box_iou
- symbol.contrib.box_nms
- symbol.contrib.box_non_maximum_suppression
- symbol.contrib.calibrate_entropy
- symbol.contrib.count_sketch
- symbol.contrib.ctc_loss
- symbol.contrib.dequantize
- symbol.contrib.dgl_adjacency
- symbol.contrib.dgl_csr_neighbor_non_uniform_sample
- symbol.contrib.dgl_csr_neighbor_uniform_sample
- symbol.contrib.dgl_graph_compact
- symbol.contrib.dgl_subgraph
- symbol.contrib.div_sqrt_dim
- symbol.contrib.edge_id
- symbol.contrib.fft
- symbol.contrib.getnnz
- symbol.contrib.gradientmultiplier
- symbol.contrib.group_adagrad_update
- symbol.contrib.hawkesll
- symbol.contrib.ifft
- symbol.contrib.index_array
- symbol.contrib.index_copy
- symbol.contrib.interleaved_matmul_encdec_qk
- symbol.contrib.interleaved_matmul_encdec_valatt
- symbol.contrib.interleaved_matmul_selfatt_qk
- symbol.contrib.interleaved_matmul_selfatt_valatt
- symbol.contrib.quadratic
- symbol.contrib.quantize
- symbol.contrib.quantize_v2
- symbol.contrib.quantized_act
- symbol.contrib.quantized_batch_norm
- symbol.contrib.quantized_concat
- symbol.contrib.quantized_conv
- symbol.contrib.quantized_elemwise_add
- symbol.contrib.quantized_elemwise_mul
- symbol.contrib.quantized_embedding
- symbol.contrib.quantized_flatten
- symbol.contrib.quantized_fully_connected
- symbol.contrib.quantized_pooling
- symbol.contrib.requantize
- symbol.contrib.round_ste
- symbol.contrib.sign_ste
-
- symbol.image
- symbol.image.adjust_lighting
- symbol.image.crop
- symbol.image.flip_left_right
- symbol.image.flip_top_bottom
- symbol.image.normalize
- symbol.image.random_brightness
- symbol.image.random_color_jitter
- symbol.image.random_contrast
- symbol.image.random_flip_left_right
- symbol.image.random_flip_top_bottom
- symbol.image.random_hue
- symbol.image.random_lighting
- symbol.image.random_saturation
- symbol.image.resize
- symbol.image.to_tensor
-
- symbol.linalg
- symbol.linalg.det
- symbol.linalg.extractdiag
- symbol.linalg.extracttrian
- symbol.linalg.gelqf
- symbol.linalg.gemm
- symbol.linalg.gemm2
- symbol.linalg.inverse
- symbol.linalg.makediag
- symbol.linalg.maketrian
- symbol.linalg.potrf
- symbol.linalg.potri
- symbol.linalg.slogdet
- symbol.linalg.sumlogdiag
- symbol.linalg.syevd
- symbol.linalg.syrk
- symbol.linalg.trmm
- symbol.linalg.trsm
-
- symbol.op
- symbol.op.Activation
- symbol.op.BatchNorm
- symbol.op.BatchNorm_v1
- symbol.op.BilinearSampler
- symbol.op.BlockGrad
- symbol.op.CTCLoss
- symbol.op.Cast
- symbol.op.Concat
- symbol.op.Convolution
- symbol.op.Convolution_v1
- symbol.op.Correlation
- symbol.op.Crop
- symbol.op.Custom
- symbol.op.Deconvolution
- symbol.op.Dropout
- symbol.op.ElementWiseSum
- symbol.op.Embedding
- symbol.op.Flatten
- symbol.op.FullyConnected
- symbol.op.GridGenerator
- symbol.op.GroupNorm
- symbol.op.IdentityAttachKLSparseReg
- symbol.op.InstanceNorm
- symbol.op.L2Normalization
- symbol.op.LRN
- symbol.op.LayerNorm
- symbol.op.LeakyReLU
- symbol.op.LinearRegressionOutput
- symbol.op.LogisticRegressionOutput
- symbol.op.MAERegressionOutput
- symbol.op.MakeLoss
- symbol.op.Pad
- symbol.op.Pooling
- symbol.op.Pooling_v1
- symbol.op.RNN
- symbol.op.ROIPooling
- symbol.op.Reshape
- symbol.op.SVMOutput
- symbol.op.SequenceLast
- symbol.op.SequenceMask
- symbol.op.SequenceReverse
- symbol.op.SliceChannel
- symbol.op.Softmax
- symbol.op.SoftmaxActivation
- symbol.op.SoftmaxOutput
- symbol.op.SpatialTransformer
- symbol.op.SwapAxis
- symbol.op.UpSampling
- symbol.op.abs
- symbol.op.adam_update
- symbol.op.add_n
- symbol.op.all_finite
- symbol.op.amp_cast
- symbol.op.amp_multicast
- symbol.op.arccos
- symbol.op.arccosh
- symbol.op.arcsin
- symbol.op.arcsinh
- symbol.op.arctan
- symbol.op.arctanh
- symbol.op.argmax
- symbol.op.argmax_channel
- symbol.op.argmin
- symbol.op.argsort
- symbol.op.batch_dot
- symbol.op.batch_take
- symbol.op.broadcast_add
- symbol.op.broadcast_axes
- symbol.op.broadcast_axis
- symbol.op.broadcast_div
- symbol.op.broadcast_equal
- symbol.op.broadcast_greater
- symbol.op.broadcast_greater_equal
- symbol.op.broadcast_hypot
- symbol.op.broadcast_lesser
- symbol.op.broadcast_lesser_equal
- symbol.op.broadcast_like
- symbol.op.broadcast_logical_and
- symbol.op.broadcast_logical_or
- symbol.op.broadcast_logical_xor
- symbol.op.broadcast_maximum
- symbol.op.broadcast_minimum
- symbol.op.broadcast_minus
- symbol.op.broadcast_mod
- symbol.op.broadcast_mul
- symbol.op.broadcast_not_equal
- symbol.op.broadcast_plus
- symbol.op.broadcast_power
- symbol.op.broadcast_sub
- symbol.op.broadcast_to
- symbol.op.cast_storage
- symbol.op.cbrt
- symbol.op.ceil
- symbol.op.choose_element_0index
- symbol.op.clip
- symbol.op.col2im
- symbol.op.cos
- symbol.op.cosh
- symbol.op.ctc_loss
- symbol.op.cumsum
- symbol.op.degrees
- symbol.op.depth_to_space
- symbol.op.diag
- symbol.op.dot
- symbol.op.elemwise_add
- symbol.op.elemwise_div
- symbol.op.elemwise_mul
- symbol.op.elemwise_sub
- symbol.op.erf
- symbol.op.erfinv
- symbol.op.exp
- symbol.op.expand_dims
- symbol.op.expm1
- symbol.op.fill_element_0index
- symbol.op.fix
- symbol.op.flip
- symbol.op.floor
- symbol.op.ftml_update
- symbol.op.ftrl_update
- symbol.op.gamma
- symbol.op.gammaln
- symbol.op.gather_nd
- symbol.op.hard_sigmoid
- symbol.op.identity
- symbol.op.im2col
- symbol.op.khatri_rao
- symbol.op.lamb_update_phase1
- symbol.op.lamb_update_phase2
- symbol.op.linalg_det
- symbol.op.linalg_extractdiag
- symbol.op.linalg_extracttrian
- symbol.op.linalg_gelqf
- symbol.op.linalg_gemm
- symbol.op.linalg_gemm2
- symbol.op.linalg_inverse
- symbol.op.linalg_makediag
- symbol.op.linalg_maketrian
- symbol.op.linalg_potrf
- symbol.op.linalg_potri
- symbol.op.linalg_slogdet
- symbol.op.linalg_sumlogdiag
- symbol.op.linalg_syrk
- symbol.op.linalg_trmm
- symbol.op.linalg_trsm
- symbol.op.log
- symbol.op.log10
- symbol.op.log1p
- symbol.op.log2
- symbol.op.log_softmax
- symbol.op.logical_not
- symbol.op.make_loss
- symbol.op.max
- symbol.op.max_axis
- symbol.op.mean
- symbol.op.min
- symbol.op.min_axis
- symbol.op.moments
- symbol.op.mp_lamb_update_phase1
- symbol.op.mp_lamb_update_phase2
- symbol.op.mp_nag_mom_update
- symbol.op.mp_sgd_mom_update
- symbol.op.mp_sgd_update
- symbol.op.multi_all_finite
- symbol.op.multi_lars
- symbol.op.multi_mp_sgd_mom_update
- symbol.op.multi_mp_sgd_update
- symbol.op.multi_sgd_mom_update
- symbol.op.multi_sgd_update
- symbol.op.multi_sum_sq
- symbol.op.nag_mom_update
- symbol.op.nanprod
- symbol.op.nansum
- symbol.op.negative
- symbol.op.norm
- symbol.op.normal
- symbol.op.one_hot
- symbol.op.ones_like
- symbol.op.pick
- symbol.op.preloaded_multi_mp_sgd_mom_update
- symbol.op.preloaded_multi_mp_sgd_update
- symbol.op.preloaded_multi_sgd_mom_update
- symbol.op.preloaded_multi_sgd_update
- symbol.op.prod
- symbol.op.radians
- symbol.op.random_exponential
- symbol.op.random_gamma
- symbol.op.random_generalized_negative_binomial
- symbol.op.random_negative_binomial
- symbol.op.random_normal
- symbol.op.random_pdf_dirichlet
- symbol.op.random_pdf_exponential
- symbol.op.random_pdf_gamma
- symbol.op.random_pdf_generalized_negative_binomial
- symbol.op.random_pdf_negative_binomial
- symbol.op.random_pdf_normal
- symbol.op.random_pdf_poisson
- symbol.op.random_pdf_uniform
- symbol.op.random_poisson
- symbol.op.random_randint
- symbol.op.random_uniform
- symbol.op.ravel_multi_index
- symbol.op.rcbrt
- symbol.op.reciprocal
- symbol.op.relu
- symbol.op.repeat
- symbol.op.reset_arrays
- symbol.op.reshape_like
- symbol.op.reverse
- symbol.op.rint
- symbol.op.rmsprop_update
- symbol.op.rmspropalex_update
- symbol.op.round
- symbol.op.rsqrt
- symbol.op.sample_exponential
- symbol.op.sample_gamma
- symbol.op.sample_generalized_negative_binomial
- symbol.op.sample_multinomial
- symbol.op.sample_negative_binomial
- symbol.op.sample_normal
- symbol.op.sample_poisson
- symbol.op.sample_uniform
- symbol.op.scatter_nd
- symbol.op.sgd_mom_update
- symbol.op.sgd_update
- symbol.op.shape_array
- symbol.op.shuffle
- symbol.op.sigmoid
- symbol.op.sign
- symbol.op.signsgd_update
- symbol.op.signum_update
- symbol.op.sin
- symbol.op.sinh
- symbol.op.size_array
- symbol.op.slice
- symbol.op.slice_axis
- symbol.op.slice_like
- symbol.op.smooth_l1
- symbol.op.softmax_cross_entropy
- symbol.op.softmin
- symbol.op.softsign
- symbol.op.sort
- symbol.op.space_to_depth
- symbol.op.split
- symbol.op.sqrt
- symbol.op.square
- symbol.op.squeeze
- symbol.op.stack
- symbol.op.stop_gradient
- symbol.op.sum
- symbol.op.sum_axis
- symbol.op.swapaxes
- symbol.op.take
- symbol.op.tan
- symbol.op.tanh
- symbol.op.tile
- symbol.op.topk
- symbol.op.transpose
- symbol.op.trunc
- symbol.op.uniform
- symbol.op.unravel_index
- symbol.op.where
- symbol.op.zeros_like
-
- symbol.random
- symbol.random.uniform
- symbol.random.normal
- symbol.random.randn
- symbol.random.poisson
- symbol.random.exponential
- symbol.random.gamma
- symbol.random.multinomial
- symbol.random.negative_binomial
- symbol.random.generalized_negative_binomial
- symbol.random.shuffle
- symbol.random.randint
- symbol.random.exponential_like
- symbol.random.gamma_like
- symbol.random.generalized_negative_binomial_like
- symbol.random.negative_binomial_like
- symbol.random.normal_like
- symbol.random.poisson_like
- symbol.random.uniform_like
- symbol.register
-
- symbol.sparse
- symbol.sparse.ElementWiseSum
- symbol.sparse.Embedding
- symbol.sparse.FullyConnected
- symbol.sparse.LinearRegressionOutput
- symbol.sparse.LogisticRegressionOutput
- symbol.sparse.MAERegressionOutput
- symbol.sparse.abs
- symbol.sparse.adagrad_update
- symbol.sparse.adam_update
- symbol.sparse.add_n
- symbol.sparse.arccos
- symbol.sparse.arccosh
- symbol.sparse.arcsin
- symbol.sparse.arcsinh
- symbol.sparse.arctan
- symbol.sparse.arctanh
- symbol.sparse.broadcast_add
- symbol.sparse.broadcast_div
- symbol.sparse.broadcast_minus
- symbol.sparse.broadcast_mul
- symbol.sparse.broadcast_plus
- symbol.sparse.broadcast_sub
- symbol.sparse.cast_storage
- symbol.sparse.cbrt
- symbol.sparse.ceil
- symbol.sparse.clip
- symbol.sparse.concat
- symbol.sparse.cos
- symbol.sparse.cosh
- symbol.sparse.degrees
- symbol.sparse.dot
- symbol.sparse.elemwise_add
- symbol.sparse.elemwise_div
- symbol.sparse.elemwise_mul
- symbol.sparse.elemwise_sub
- symbol.sparse.exp
- symbol.sparse.expm1
- symbol.sparse.fix
- symbol.sparse.floor
- symbol.sparse.ftrl_update
- symbol.sparse.gamma
- symbol.sparse.gammaln
- symbol.sparse.log
- symbol.sparse.log10
- symbol.sparse.log1p
- symbol.sparse.log2
- symbol.sparse.make_loss
- symbol.sparse.mean
- symbol.sparse.negative
- symbol.sparse.norm
- symbol.sparse.radians
- symbol.sparse.relu
- symbol.sparse.retain
- symbol.sparse.rint
- symbol.sparse.round
- symbol.sparse.rsqrt
- symbol.sparse.sgd_mom_update
- symbol.sparse.sgd_update
- symbol.sparse.sigmoid
- symbol.sparse.sign
- symbol.sparse.sin
- symbol.sparse.sinh
- symbol.sparse.slice
- symbol.sparse.sqrt
- symbol.sparse.square
- symbol.sparse.stop_gradient
- symbol.sparse.sum
- symbol.sparse.tan
- symbol.sparse.tanh
- symbol.sparse.trunc
- symbol.sparse.where
- symbol.sparse.zeros_like
- symbol.Activation
- symbol.BatchNorm
- symbol.BatchNorm_v1
- symbol.BilinearSampler
- symbol.BlockGrad
- symbol.CTCLoss
- symbol.Cast
- symbol.Concat
- symbol.Convolution
- symbol.Convolution_v1
- symbol.Correlation
- symbol.Crop
- symbol.Custom
- symbol.Deconvolution
- symbol.Dropout
- symbol.ElementWiseSum
- symbol.Embedding
- symbol.Flatten
- symbol.FullyConnected
- symbol.GridGenerator
- symbol.GroupNorm
- symbol.IdentityAttachKLSparseReg
- symbol.InstanceNorm
- symbol.L2Normalization
- symbol.LRN
- symbol.LayerNorm
- symbol.LeakyReLU
- symbol.LinearRegressionOutput
- symbol.LogisticRegressionOutput
- symbol.MAERegressionOutput
- symbol.MakeLoss
- symbol.Pad
- symbol.Pooling
- symbol.Pooling_v1
- symbol.RNN
- symbol.ROIPooling
- symbol.Reshape
- symbol.SVMOutput
- symbol.SequenceLast
- symbol.SequenceMask
- symbol.SequenceReverse
- symbol.SliceChannel
- symbol.Softmax
- symbol.SoftmaxActivation
- symbol.SoftmaxOutput
- symbol.SpatialTransformer
- symbol.SwapAxis
- symbol.UpSampling
- symbol.abs
- symbol.adam_update
- symbol.add_n
- symbol.all_finite
- symbol.amp_cast
- symbol.amp_multicast
- symbol.arccos
- symbol.arccosh
- symbol.arcsin
- symbol.arcsinh
- symbol.arctan
- symbol.arctanh
- symbol.argmax
- symbol.argmax_channel
- symbol.argmin
- symbol.argsort
- symbol.batch_dot
- symbol.batch_take
- symbol.broadcast_add
- symbol.broadcast_axes
- symbol.broadcast_axis
- symbol.broadcast_div
- symbol.broadcast_equal
- symbol.broadcast_greater
- symbol.broadcast_greater_equal
- symbol.broadcast_hypot
- symbol.broadcast_lesser
- symbol.broadcast_lesser_equal
- symbol.broadcast_like
- symbol.broadcast_logical_and
- symbol.broadcast_logical_or
- symbol.broadcast_logical_xor
- symbol.broadcast_maximum
- symbol.broadcast_minimum
- symbol.broadcast_minus
- symbol.broadcast_mod
- symbol.broadcast_mul
- symbol.broadcast_not_equal
- symbol.broadcast_plus
- symbol.broadcast_power
- symbol.broadcast_sub
- symbol.broadcast_to
- symbol.cast_storage
- symbol.cbrt
- symbol.ceil
- symbol.choose_element_0index
- symbol.clip
- symbol.col2im
- symbol.cos
- symbol.cosh
- symbol.ctc_loss
- symbol.cumsum
- symbol.degrees
- symbol.depth_to_space
- symbol.diag
- symbol.dot
- symbol.elemwise_add
- symbol.elemwise_div
- symbol.elemwise_mul
- symbol.elemwise_sub
- symbol.erf
- symbol.erfinv
- symbol.exp
- symbol.expand_dims
- symbol.expm1
- symbol.fill_element_0index
- symbol.fix
- symbol.flip
- symbol.floor
- symbol.ftml_update
- symbol.ftrl_update
- symbol.gamma
- symbol.gammaln
- symbol.gather_nd
- symbol.hard_sigmoid
- symbol.identity
- symbol.im2col
- symbol.khatri_rao
- symbol.lamb_update_phase1
- symbol.lamb_update_phase2
- symbol.linalg_det
- symbol.linalg_extractdiag
- symbol.linalg_extracttrian
- symbol.linalg_gelqf
- symbol.linalg_gemm
- symbol.linalg_gemm2
- symbol.linalg_inverse
- symbol.linalg_makediag
- symbol.linalg_maketrian
- symbol.linalg_potrf
- symbol.linalg_potri
- symbol.linalg_slogdet
- symbol.linalg_sumlogdiag
- symbol.linalg_syrk
- symbol.linalg_trmm
- symbol.linalg_trsm
- symbol.log
- symbol.log10
- symbol.log1p
- symbol.log2
- symbol.log_softmax
- symbol.logical_not
- symbol.make_loss
- symbol.max
- symbol.max_axis
- symbol.mean
- symbol.min
- symbol.min_axis
- symbol.moments
- symbol.mp_lamb_update_phase1
- symbol.mp_lamb_update_phase2
- symbol.mp_nag_mom_update
- symbol.mp_sgd_mom_update
- symbol.mp_sgd_update
- symbol.multi_all_finite
- symbol.multi_lars
- symbol.multi_mp_sgd_mom_update
- symbol.multi_mp_sgd_update
- symbol.multi_sgd_mom_update
- symbol.multi_sgd_update
- symbol.multi_sum_sq
- symbol.nag_mom_update
- symbol.nanprod
- symbol.nansum
- symbol.negative
- symbol.norm
- symbol.normal
- symbol.one_hot
- symbol.ones_like
- symbol.pick
- symbol.preloaded_multi_mp_sgd_mom_update
- symbol.preloaded_multi_mp_sgd_update
- symbol.preloaded_multi_sgd_mom_update
- symbol.preloaded_multi_sgd_update
- symbol.prod
- symbol.radians
- symbol.random_exponential
- symbol.random_gamma
- symbol.random_generalized_negative_binomial
- symbol.random_negative_binomial
- symbol.random_normal
- symbol.random_pdf_dirichlet
- symbol.random_pdf_exponential
- symbol.random_pdf_gamma
- symbol.random_pdf_generalized_negative_binomial
- symbol.random_pdf_negative_binomial
- symbol.random_pdf_normal
- symbol.random_pdf_poisson
- symbol.random_pdf_uniform
- symbol.random_poisson
- symbol.random_randint
- symbol.random_uniform
- symbol.ravel_multi_index
- symbol.rcbrt
- symbol.reciprocal
- symbol.relu
- symbol.repeat
- symbol.reset_arrays
- symbol.reshape_like
- symbol.reverse
- symbol.rint
- symbol.rmsprop_update
- symbol.rmspropalex_update
- symbol.round
- symbol.rsqrt
- symbol.sample_exponential
- symbol.sample_gamma
- symbol.sample_generalized_negative_binomial
- symbol.sample_multinomial
- symbol.sample_negative_binomial
- symbol.sample_normal
- symbol.sample_poisson
- symbol.sample_uniform
- symbol.scatter_nd
- symbol.sgd_mom_update
- symbol.sgd_update
- symbol.shape_array
- symbol.shuffle
- symbol.sigmoid
- symbol.sign
- symbol.signsgd_update
- symbol.signum_update
- symbol.sin
- symbol.sinh
- symbol.size_array
- symbol.slice
- symbol.slice_axis
- symbol.slice_like
- symbol.smooth_l1
- symbol.softmax_cross_entropy
- symbol.softmin
- symbol.softsign
- symbol.sort
- symbol.space_to_depth
- symbol.split
- symbol.sqrt
- symbol.square
- symbol.squeeze
- symbol.stack
- symbol.stop_gradient
- symbol.sum
- symbol.sum_axis
- symbol.swapaxes
- symbol.take
- symbol.tan
- symbol.tanh
- symbol.tile
- symbol.topk
- symbol.transpose
- symbol.trunc
- symbol.uniform
- symbol.unravel_index
- symbol.where
- symbol.zeros_like
- symbol.var
- symbol.Variable
- symbol.Group
- symbol.load
- symbol.load_json
- symbol.pow
- symbol.power
- symbol.maximum
- symbol.minimum
- symbol.hypot
- symbol.eye
- symbol.zeros
- symbol.ones
- symbol.full
- symbol.arange
- symbol.linspace
- symbol.histogram
- symbol.split_v2
-
- contrib.ndarray
- contrib.ndarray.AdaptiveAvgPooling2D
- contrib.ndarray.BilinearResize2D
- contrib.ndarray.CTCLoss
- contrib.ndarray.DeformableConvolution
- contrib.ndarray.DeformablePSROIPooling
- contrib.ndarray.ModulatedDeformableConvolution
- contrib.ndarray.MultiBoxDetection
- contrib.ndarray.MultiBoxPrior
- contrib.ndarray.MultiBoxTarget
- contrib.ndarray.MultiProposal
- contrib.ndarray.PSROIPooling
- contrib.ndarray.Proposal
- contrib.ndarray.ROIAlign
- contrib.ndarray.RROIAlign
- contrib.ndarray.SparseEmbedding
- contrib.ndarray.SyncBatchNorm
- contrib.ndarray.allclose
- contrib.ndarray.arange_like
- contrib.ndarray.backward_gradientmultiplier
- contrib.ndarray.backward_hawkesll
- contrib.ndarray.backward_index_copy
- contrib.ndarray.backward_quadratic
- contrib.ndarray.bipartite_matching
- contrib.ndarray.boolean_mask
- contrib.ndarray.box_decode
- contrib.ndarray.box_encode
- contrib.ndarray.box_iou
- contrib.ndarray.box_nms
- contrib.ndarray.box_non_maximum_suppression
- contrib.ndarray.calibrate_entropy
- contrib.ndarray.count_sketch
- contrib.ndarray.ctc_loss
- contrib.ndarray.dequantize
- contrib.ndarray.dgl_adjacency
- contrib.ndarray.dgl_csr_neighbor_non_uniform_sample
- contrib.ndarray.dgl_csr_neighbor_uniform_sample
- contrib.ndarray.dgl_graph_compact
- contrib.ndarray.dgl_subgraph
- contrib.ndarray.div_sqrt_dim
- contrib.ndarray.edge_id
- contrib.ndarray.fft
- contrib.ndarray.getnnz
- contrib.ndarray.gradientmultiplier
- contrib.ndarray.group_adagrad_update
- contrib.ndarray.hawkesll
- contrib.ndarray.ifft
- contrib.ndarray.index_array
- contrib.ndarray.index_copy
- contrib.ndarray.interleaved_matmul_encdec_qk
- contrib.ndarray.interleaved_matmul_encdec_valatt
- contrib.ndarray.interleaved_matmul_selfatt_qk
- contrib.ndarray.interleaved_matmul_selfatt_valatt
- contrib.ndarray.quadratic
- contrib.ndarray.quantize
- contrib.ndarray.quantize_v2
- contrib.ndarray.quantized_act
- contrib.ndarray.quantized_batch_norm
- contrib.ndarray.quantized_concat
- contrib.ndarray.quantized_conv
- contrib.ndarray.quantized_elemwise_add
- contrib.ndarray.quantized_elemwise_mul
- contrib.ndarray.quantized_embedding
- contrib.ndarray.quantized_flatten
- contrib.ndarray.quantized_fully_connected
- contrib.ndarray.quantized_pooling
- contrib.ndarray.requantize
- contrib.ndarray.round_ste
- contrib.ndarray.sign_ste
-
- contrib.symbol
- contrib.symbol.AdaptiveAvgPooling2D
- contrib.symbol.BilinearResize2D
- contrib.symbol.CTCLoss
- contrib.symbol.DeformableConvolution
- contrib.symbol.DeformablePSROIPooling
- contrib.symbol.ModulatedDeformableConvolution
- contrib.symbol.MultiBoxDetection
- contrib.symbol.MultiBoxPrior
- contrib.symbol.MultiBoxTarget
- contrib.symbol.MultiProposal
- contrib.symbol.PSROIPooling
- contrib.symbol.Proposal
- contrib.symbol.ROIAlign
- contrib.symbol.RROIAlign
- contrib.symbol.SparseEmbedding
- contrib.symbol.SyncBatchNorm
- contrib.symbol.allclose
- contrib.symbol.arange_like
- contrib.symbol.backward_gradientmultiplier
- contrib.symbol.backward_hawkesll
- contrib.symbol.backward_index_copy
- contrib.symbol.backward_quadratic
- contrib.symbol.bipartite_matching
- contrib.symbol.boolean_mask
- contrib.symbol.box_decode
- contrib.symbol.box_encode
- contrib.symbol.box_iou
- contrib.symbol.box_nms
- contrib.symbol.box_non_maximum_suppression
- contrib.symbol.calibrate_entropy
- contrib.symbol.count_sketch
- contrib.symbol.ctc_loss
- contrib.symbol.dequantize
- contrib.symbol.dgl_adjacency
- contrib.symbol.dgl_csr_neighbor_non_uniform_sample
- contrib.symbol.dgl_csr_neighbor_uniform_sample
- contrib.symbol.dgl_graph_compact
- contrib.symbol.dgl_subgraph
- contrib.symbol.div_sqrt_dim
- contrib.symbol.edge_id
- contrib.symbol.fft
- contrib.symbol.getnnz
- contrib.symbol.gradientmultiplier
- contrib.symbol.group_adagrad_update
- contrib.symbol.hawkesll
- contrib.symbol.ifft
- contrib.symbol.index_array
- contrib.symbol.index_copy
- contrib.symbol.interleaved_matmul_encdec_qk
- contrib.symbol.interleaved_matmul_encdec_valatt
- contrib.symbol.interleaved_matmul_selfatt_qk
- contrib.symbol.interleaved_matmul_selfatt_valatt
- contrib.symbol.quadratic
- contrib.symbol.quantize
- contrib.symbol.quantize_v2
- contrib.symbol.quantized_act
- contrib.symbol.quantized_batch_norm
- contrib.symbol.quantized_concat
- contrib.symbol.quantized_conv
- contrib.symbol.quantized_elemwise_add
- contrib.symbol.quantized_elemwise_mul
- contrib.symbol.quantized_embedding
- contrib.symbol.quantized_flatten
- contrib.symbol.quantized_fully_connected
- contrib.symbol.quantized_pooling
- contrib.symbol.requantize
- contrib.symbol.round_ste
- contrib.symbol.sign_ste
- contrib.text
- mxnet.attribute
- mxnet.base
- mxnet.callback
- mxnet.context
- mxnet.engine
- mxnet.executor
- mxnet.executor_manager
- mxnet.image
- mxnet.io
- mxnet.kvstore_server
- mxnet.libinfo
- mxnet.log
- mxnet.model
- mxnet.monitor
- mxnet.name
- mxnet.notebook
- mxnet.operator
- mxnet.profiler
- mxnet.random
- mxnet.recordio
- mxnet.registry
- mxnet.rtc
- mxnet.runtime
- mxnet.test_utils
- mxnet.torch
- mxnet.util
- mxnet.visualization
mxnet / mxnet.test_utils
mxnet.test_utils¶
Tools for testing.
Classes
|
A dummy iterator that always returns the same batch of data (the first data batch of the real data iter). |
|
Environment variable setter and unsetter via with idiom |
Functions
|
Test if two numpy arrays are almost equal. |
|
Test that two NumPy arrays are almost equal (ignoring NaN in either array). |
|
Test that two numpy arrays are almost equal. |
|
Test that two NumPy arrays are almost equal (ignoring NaN in either array). |
|
Test that two numpy arrays are almost equal within given error rate. |
|
Test that function f will throw an exception of type given by exception_type |
|
Return ndarray composed of passing each array value through some function |
|
Return ndarray composed of passing two array values through some function |
|
Check symbol gives the same output for different running context |
Check whether a HybridBlock has consistent output between the hybridized v.s. |
|
|
Verify an operation by checking backward pass via finite difference method. |
|
Check the running speed of a symbol. |
|
Compares a symbol’s backward results with the expected ones. |
|
Compares a symbol’s forward results with the expected ones. |
|
Run the chi-square test for the generator. |
|
Given a as a numpy ndarray, perform reduce_sum on a over the axes that do not exist in shape. |
|
Compare ndarray tuple. |
|
Compare opt1 and opt2. |
|
Create a sparse array, For Rsp, assure indices are in a canonical format |
|
Create sparse array, using only rsp_indices to determine density |
Get default context for regression test. |
|
Get default data type for regression test. |
|
Discards error output of a routine if invoked as: |
|
|
Download an given URL |
|
Download a model from data.mxnet.io |
|
Finds and returns the location of maximum violation. |
|
Generate the buckets and probabilities for chi_square test when the ppf (Quantile function) is specified. |
|
Get default numerical threshold for regression test. |
|
Download and extract bz2 data. |
Downloads CIFAR10 dataset into a directory in the current directory with the name data, and then extracts all files into the directory data/cifar. |
|
|
Get default numerical threshold for regression test. |
|
Get path to the im2rec.py tool |
Download and load the MNIST dataset |
|
|
Returns training and validation iterators for MNIST dataset |
Downloads MNIST dataset as a pkl.gz into a directory in the current directory |
|
Downloads ubyte version of the MNIST dataset into a directory in the current directory with the name data and extracts all files in the zip archive to this directory. |
|
|
Get default numerical threshold for regression test. |
|
Download and extract zip data. |
Returns True if MXNet is compiled with TVM generated operators. |
|
Checks if the test is running as part of a Continuous Delivery run |
|
Returns True for all CPU tests. |
|
Return a list of GPUs |
|
|
Create element mismatch comment |
|
Test the generator by matching the mean. |
Generate a well-conditioned matrix with small real eigenvalues. |
|
Generate well-conditioned matrices with small real eigenvalues. |
|
Generate a orthonormal matrix. |
|
Generate a sym matrix with real eigenvalues. |
|
Generate sym matrices with real eigenvalues. |
|
|
Compatible reduce for old version of NumPy. |
|
Calculates a numeric gradient via finite difference method. |
|
Generate a random sparse ndarray. |
|
Generate a random sparse ndarray. |
|
Generate some random numpy arrays. |
|
Return a k length list of the elements chosen from the population sequence. |
|
Generate some random numpy arrays. |
|
Retry n times before failing for stochastic test cases. |
|
Test if two NumPy arrays are the same. |
|
Check whether two NDArrays sharing the same memory block |
|
Compare two symbols to check if they have the same computation graph structure. |
|
Set default context. |
|
Set environment variable |
Shuffle CSR column indices per row |
|
|
A simple forward function for a symbol. |
|
Test the generator by matching the variance. |
|
Verify whether the generator is correct using chi-square testing. |
-
class
mxnet.test_utils.
DummyIter
(real_iter)[source]¶ Bases:
mxnet.io.io.DataIter
A dummy iterator that always returns the same batch of data (the first data batch of the real data iter). This is usually used for speed testing.
- Parameters
real_iter (mx.io.DataIter) – The real data iterator where the first batch of data comes from
Methods
next
()Get a data batch from iterator.
-
class
mxnet.test_utils.
EnvManager
(key, val)[source]¶ Bases:
object
Environment variable setter and unsetter via with idiom
-
mxnet.test_utils.
almost_equal
(a, b, rtol=None, atol=None, equal_nan=False, use_broadcast=True)[source]¶ Test if two numpy arrays are almost equal.
-
mxnet.test_utils.
almost_equal_ignore_nan
(a, b, rtol=None, atol=None)[source]¶ Test that two NumPy arrays are almost equal (ignoring NaN in either array). Combines a relative and absolute measure of approximate eqality. If either the relative or absolute check passes, the arrays are considered equal. Including an absolute check resolves issues with the relative check where all array values are close to zero.
- Parameters
a (np.ndarray) –
b (np.ndarray) –
rtol (None or float) – The relative threshold. Default threshold will be used if set to
None
.atol (None or float) – The absolute threshold. Default threshold will be used if set to
None
.
-
mxnet.test_utils.
assert_almost_equal
(a, b, rtol=None, atol=None, names=('a', 'b'), equal_nan=False, use_broadcast=True, mismatches=(10, 10))[source]¶ Test that two numpy arrays are almost equal. Raise exception message if not.
- Parameters
a (np.ndarray or mx.nd.array) –
b (np.ndarray or mx.nd.array) –
rtol (None or float) – The relative threshold. Default threshold will be used if set to
None
.atol (None or float) – The absolute threshold. Default threshold will be used if set to
None
.names (tuple of names, optional) – The names used in error message when an exception occurs
equal_nan (boolean, optional) – The flag determining how to treat NAN values in comparison
mismatches (tuple of mismatches) – Maximum number of mismatches to be printed (mismatches[0]) and determine (mismatches[1])
-
mxnet.test_utils.
assert_almost_equal_ignore_nan
(a, b, rtol=None, atol=None, names=('a', 'b'))[source]¶ Test that two NumPy arrays are almost equal (ignoring NaN in either array). Combines a relative and absolute measure of approximate eqality. If either the relative or absolute check passes, the arrays are considered equal. Including an absolute check resolves issues with the relative check where all array values are close to zero.
- Parameters
a (np.ndarray) –
b (np.ndarray) –
rtol (None or float) – The relative threshold. Default threshold will be used if set to
None
.atol (None or float) – The absolute threshold. Default threshold will be used if set to
None
.
-
mxnet.test_utils.
assert_almost_equal_with_err
(a, b, rtol=None, atol=None, etol=None, names=('a', 'b'), equal_nan=False, mismatches=(10, 10))[source]¶ Test that two numpy arrays are almost equal within given error rate. Raise exception message if not.
- Parameters
a (np.ndarray) –
b (np.ndarray) –
threshold (None or float) – The checking threshold. Default threshold will be used if set to
None
.etol (None or float) – The error rate threshold. If etol is float, return true if error_rate < etol even if any error is found.
-
mxnet.test_utils.
assert_exception
(f, exception_type, *args, **kwargs)[source]¶ Test that function f will throw an exception of type given by exception_type
-
mxnet.test_utils.
assign_each
(the_input, function)[source]¶ Return ndarray composed of passing each array value through some function
-
mxnet.test_utils.
assign_each2
(input1, input2, function)[source]¶ Return ndarray composed of passing two array values through some function
-
mxnet.test_utils.
check_consistency
(sym, ctx_list, scale=1.0, grad_req='write', arg_params=None, aux_params=None, tol=None, raise_on_err=True, ground_truth=None, equal_nan=False, use_uniform=False, rand_type=<class 'numpy.float64'>)[source]¶ Check symbol gives the same output for different running context
- Parameters
sym (Symbol or list of Symbols) – Symbol(s) to run the consistency test.
ctx_list (list) – Running context. See example for more detail.
scale (float, optional) – Standard deviation of the inner normal distribution. Used in initialization.
grad_req (str or list of str or dict of str to str) – Gradient requirement.
use_unifrom (bool) – Optional, When flag set to true, random input data generated follows uniform distribution, not normal distribution
rand_type (np.dtype) – casts the randomly generated data to this type Optional, when input data is passed via arg_params, defaults to np.float64 (numpy float default)
Examples
>>> # create the symbol >>> sym = mx.sym.Convolution(num_filter=3, kernel=(3,3), name='conv') >>> # initialize the running context >>> ctx_list =[{'ctx': mx.gpu(0), 'conv_data': (2, 2, 10, 10), 'type_dict': {'conv_data': np.float64}}, {'ctx': mx.gpu(0), 'conv_data': (2, 2, 10, 10), 'type_dict': {'conv_data': np.float32}}, {'ctx': mx.gpu(0), 'conv_data': (2, 2, 10, 10), 'type_dict': {'conv_data': np.float16}}, {'ctx': mx.cpu(0), 'conv_data': (2, 2, 10, 10), 'type_dict': {'conv_data': np.float64}}, {'ctx': mx.cpu(0), 'conv_data': (2, 2, 10, 10), 'type_dict': {'conv_data': np.float32}}] >>> check_consistency(sym, ctx_list) >>> sym = mx.sym.Concat(name='concat', num_args=2) >>> ctx_list = [{'ctx': mx.gpu(0), 'concat_arg1': (2, 10), 'concat_arg0': (2, 10), 'type_dict': {'concat_arg0': np.float64, 'concat_arg1': np.float64}}, {'ctx': mx.gpu(0), 'concat_arg1': (2, 10), 'concat_arg0': (2, 10), 'type_dict': {'concat_arg0': np.float32, 'concat_arg1': np.float32}}, {'ctx': mx.gpu(0), 'concat_arg1': (2, 10), 'concat_arg0': (2, 10), 'type_dict': {'concat_arg0': np.float16, 'concat_arg1': np.float16}}, {'ctx': mx.cpu(0), 'concat_arg1': (2, 10), 'concat_arg0': (2, 10), 'type_dict': {'concat_arg0': np.float64, 'concat_arg1': np.float64}}, {'ctx': mx.cpu(0), 'concat_arg1': (2, 10), 'concat_arg0': (2, 10), 'type_dict': {'concat_arg0': np.float32, 'concat_arg1': np.float32}}] >>> check_consistency(sym, ctx_list)
-
mxnet.test_utils.
check_gluon_hybridize_consistency
(net_builder, data_l, numpy_func=None, test_grad=True, rtol=0.0001, atol=0.0001)[source]¶ - Check whether a HybridBlock has consistent output between the hybridized
v.s. non-hybridized versions
The network should not contain any random number generators.
- Parameters
net_builder (function) – The builder of the HybridBlock that we are going to check the consistency. Inside the implementation, we will call net_builder() to construct the hybrid block. Also, the net_builder will need to support specifying the params
data_l (list of mx.np.ndarray) – List of input ndarrays.
numpy_func (function, optional) – The ground truth numpy function that has the same functionality as net_builder(). Default None.
test_grad (bool, optional) – Whether to test the consistency of the gradient. Default True.
rtol (float, optional) – The relative error tolerance, default 1E-4. Default 1E-4.
atol (float, optional) – The absolute error tolerance, default 1E-4. Default 1E-4.
-
mxnet.test_utils.
check_numeric_gradient
(sym, location, aux_states=None, numeric_eps=0.001, rtol=0.01, atol=None, grad_nodes=None, use_forward_train=True, ctx=None, grad_stype_dict=None, dtype=<class 'numpy.float32'>)[source]¶ Verify an operation by checking backward pass via finite difference method.
Based on Theano’s theano.gradient.verify_grad [1]
- Parameters
sym (Symbol) – Symbol containing op to test
location (list or tuple or dict) –
Argument values used as location to compute gradient
if type is list of numpy.ndarray, inner elements should have the same order as mxnet.sym.list_arguments().
if type is dict of str -> numpy.ndarray, maps the name of arguments to the corresponding numpy.ndarray.
In either case, value of all the arguments must be provided.
aux_states (list or tuple or dict, optional) – The auxiliary states required when generating the executor for the symbol.
numeric_eps (float, optional) – Delta for the finite difference method that approximates the gradient.
check_eps (float, optional) – relative error eps used when comparing numeric grad to symbolic grad.
grad_nodes (None or list or tuple or dict, optional) – Names of the nodes to check gradient on
use_forward_train (bool) – Whether to use is_train=True when computing the finite-difference.
ctx (Context, optional) – Check the gradient computation on the specified device.
grad_stype_dict (dict of str->str, optional) – Storage type dictionary for gradient ndarrays.
dtype (np.float16 or np.float32 or np.float64) – Datatype for mx.nd.array.
References
[1] https://github.com/Theano/Theano/blob/master/theano/gradient.py
-
mxnet.test_utils.
check_speed
(sym, location=None, ctx=None, N=20, grad_req=None, typ='whole', **kwargs)[source]¶ Check the running speed of a symbol.
- Parameters
sym (Symbol) – Symbol to run the speed test.
location (none or dict of str to np.ndarray) – Location to evaluate the inner executor.
ctx (Context) – Running context.
N (int, optional) – Repeat times.
grad_req (None or str or list of str or dict of str to str, optional) – Gradient requirements.
typ (str, optional) –
“whole” or “forward”
- ”whole”
Test the forward_backward speed.
- ”forward”
Only test the forward speed.
-
mxnet.test_utils.
check_symbolic_backward
(sym, location, out_grads, expected, rtol=1e-05, atol=None, aux_states=None, grad_req='write', ctx=None, grad_stypes=None, equal_nan=False, dtype=<class 'numpy.float32'>)[source]¶ Compares a symbol’s backward results with the expected ones. Prints error messages if the backward results are not the same as the expected results.
- Parameters
sym (Symbol) – output symbol
location (list of np.ndarray or dict of str to np.ndarray) –
The evaluation point
- if type is list of np.ndarray
Contains all the NumPy arrays corresponding to
mx.sym.list_arguments
.
- if type is dict of str to np.ndarray
Contains the mapping between argument names and their values.
out_grads (None or list of np.ndarray or dict of str to np.ndarray) –
NumPys arrays corresponding to sym.outputs for incomming gradient.
- if type is list of np.ndarray
Contains arrays corresponding to
exe.outputs
.
- if type is dict of str to np.ndarray
contains mapping between mxnet.sym.list_output() and Executor.outputs
expected (list of np.ndarray or dict of str to np.ndarray) –
expected gradient values
- if type is list of np.ndarray
Contains arrays corresponding to exe.grad_arrays
- if type is dict of str to np.ndarray
Contains mapping between
sym.list_arguments()
and exe.outputs.
check_eps (float, optional) – Relative error to check to.
aux_states (list of np.ndarray or dict of str to np.ndarray) –
grad_req (str or list of str or dict of str to str, optional) – Gradient requirements. ‘write’, ‘add’ or ‘null’.
ctx (Context, optional) – Running context.
grad_stypes (dict of str->str) – dictionary of mapping argument name to stype for the gradient
equal_nan (Boolean) – if True, nan is a valid value for checking equivalency (ie nan == nan)
dtype (np.float16 or np.float32 or np.float64) – Datatype for mx.nd.array.
Example
>>> lhs = mx.symbol.Variable('lhs') >>> rhs = mx.symbol.Variable('rhs') >>> sym_add = mx.symbol.elemwise_add(lhs, rhs) >>> mat1 = np.array([[1, 2], [3, 4]]) >>> mat2 = np.array([[5, 6], [7, 8]]) >>> grad1 = mx.nd.zeros(shape) >>> grad2 = mx.nd.zeros(shape) >>> exec_add = sym_add.bind(default_context(), args={'lhs': mat1, 'rhs': mat2}, ... args_grad={'lhs': grad1, 'rhs': grad2}, grad_req={'lhs': 'write', 'rhs': 'write'}) >>> exec_add.forward(is_train=True) >>> ograd = mx.nd.ones(shape) >>> grad_expected = ograd.copy().asnumpy() >>> check_symbolic_backward(sym_add, [mat1, mat2], [ograd], [grad_expected, grad_expected])
-
mxnet.test_utils.
check_symbolic_forward
(sym, location, expected, rtol=0.0001, atol=None, aux_states=None, ctx=None, equal_nan=False, dtype=<class 'numpy.float32'>)[source]¶ Compares a symbol’s forward results with the expected ones. Prints error messages if the forward results are not the same as the expected ones.
- Parameters
sym (Symbol) – output symbol
location (list of np.ndarray or dict of str to np.ndarray) –
The evaluation point
- if type is list of np.ndarray
Contains all the numpy arrays corresponding to sym.list_arguments().
- if type is dict of str to np.ndarray
Contains the mapping between argument names and their values.
expected (list of np.ndarray or dict of str to np.ndarray) –
The expected output value
- if type is list of np.ndarray
Contains arrays corresponding to exe.outputs.
- if type is dict of str to np.ndarray
Contains mapping between sym.list_output() and exe.outputs.
check_eps (float, optional) – Relative error to check to.
aux_states (list of np.ndarray of dict, optional) –
- if type is list of np.ndarray
Contains all the NumPy arrays corresponding to sym.list_auxiliary_states
- if type is dict of str to np.ndarray
Contains the mapping between names of auxiliary states and their values.
ctx (Context, optional) – running context
dtype ("asnumpy" or np.float16 or np.float32 or np.float64) – If dtype is “asnumpy” then the mx.nd.array created will have the same type as th numpy array from which it is copied. Otherwise, dtype is the explicit datatype for all mx.nd.array objects created in this function.
equal_nan (Boolean) – if True, nan is a valid value for checking equivalency (ie nan == nan)
Example
>>> shape = (2, 2) >>> lhs = mx.symbol.Variable('lhs') >>> rhs = mx.symbol.Variable('rhs') >>> sym_dot = mx.symbol.dot(lhs, rhs) >>> mat1 = np.array([[1, 2], [3, 4]]) >>> mat2 = np.array([[5, 6], [7, 8]]) >>> ret_expected = np.array([[19, 22], [43, 50]]) >>> check_symbolic_forward(sym_dot, [mat1, mat2], [ret_expected])
-
mxnet.test_utils.
chi_square_check
(generator, buckets, probs, nsamples=1000000)[source]¶ Run the chi-square test for the generator. The generator can be both continuous and discrete.
If the generator is continuous, the buckets should contain tuples of (range_min, range_max) and the probs should be the corresponding ideal probability within the specific ranges. Otherwise, the buckets should contain all the possible values generated over the discrete distribution and the probs should be groud-truth probability.
Usually the user is required to specify the probs parameter.
After obtaining the p value, we could further use the standard p > 0.05 (alpha) threshold to get the final result.
Examples:
buckets, probs = gen_buckets_probs_with_ppf(lambda x: ss.norm.ppf(x, 0, 1), 5) generator = lambda x: np.random.normal(0, 1.0, size=x) p = chi_square_check(generator=generator, buckets=buckets, probs=probs) assert(p > 0.05)
- Parameters
generator (function) – A function that is assumed to generate i.i.d samples from a specific distribution. generator(N) should generate N random samples.
buckets (list of tuple or list of number) – The buckets to run the chi-square the test. Make sure that the buckets cover the whole range of the distribution. Also, the buckets must be in ascending order and have no intersection
probs (list or tuple) – The ground-truth probability of the random value fall in a specific bucket.
nsamples (int) – The number of samples to generate for the testing
- Returns
p (float) – p value that the generator has the expected distribution. A higher value indicates a larger confidence
obs_freq (list) – Observed frequency of buckets
expected_freq (list) – The expected (ground-truth) frequency of the buckets
-
mxnet.test_utils.
collapse_sum_like
(a, shape)[source]¶ Given a as a numpy ndarray, perform reduce_sum on a over the axes that do not exist in shape. Note that an ndarray with shape must be broadcastable to a.
-
mxnet.test_utils.
compare_ndarray_tuple
(t1, t2, rtol=None, atol=None)[source]¶ Compare ndarray tuple.
-
mxnet.test_utils.
compare_optimizer
(opt1, opt2, shape, dtype, w_stype='default', g_stype='default', rtol=0.0001, atol=1e-05, compare_states=True, ntensors=1)[source]¶ Compare opt1 and opt2.
-
mxnet.test_utils.
create_sparse_array
(shape, stype, data_init=None, rsp_indices=None, dtype=None, modifier_func=None, density=0.5, shuffle_csr_indices=False)[source]¶ Create a sparse array, For Rsp, assure indices are in a canonical format
-
mxnet.test_utils.
create_sparse_array_zd
(shape, stype, density, data_init=None, rsp_indices=None, dtype=None, modifier_func=None, shuffle_csr_indices=False)[source]¶ Create sparse array, using only rsp_indices to determine density
-
mxnet.test_utils.
discard_stderr
()[source]¶ Discards error output of a routine if invoked as:
- with discard_stderr():
…
-
mxnet.test_utils.
download
(url, fname=None, dirname=None, overwrite=False, retries=5)[source]¶ Download an given URL
- Parameters
url (str) – URL to download
fname (str, optional) – filename of the downloaded file. If None, then will guess a filename from url.
dirname (str, optional) – output directory name. If None, then guess from fname or use the current directory
overwrite (bool, optional) – Default is false, which means skipping download if the local file exists. If true, then download the url to overwrite the local file if exists.
retries (integer, default 5) – The number of times to attempt the download in case of failure or non 200 return codes
- Returns
The filename of the downloaded file
- Return type
str
-
mxnet.test_utils.
download_model
(model_name, dst_dir='./', meta_info=None)[source]¶ Download a model from data.mxnet.io
- Parameters
model_name (str) – Model name to download
dst_dir (str) – Destination Directory to download the model
meta_info (dict of dict) – Mapping from model_name to dict of the following structure: {‘symbol’: url, ‘params’: url}
- Returns
- Return type
Two element tuple containing model_name and epoch for the params saved
-
mxnet.test_utils.
find_max_violation
(a, b, rtol=None, atol=None)[source]¶ Finds and returns the location of maximum violation.
-
mxnet.test_utils.
gen_buckets_probs_with_ppf
(ppf, nbuckets)[source]¶ - Generate the buckets and probabilities for chi_square test when the ppf (Quantile function)
is specified.
- Parameters
ppf (function) – The Quantile function that takes a probability and maps it back to a value. It’s the inverse of the cdf function
nbuckets (int) – size of the buckets
- Returns
buckets (list of tuple) – The generated buckets
probs (list) – The generate probabilities
-
mxnet.test_utils.
get_bz2_data
(data_dir, data_name, url, data_origin_name)[source]¶ Download and extract bz2 data.
- Parameters
data_dir (str) – Absolute or relative path of the directory name to store bz2 files
data_name (str) – Name of the output file in which bz2 contents will be extracted
url (str) – URL to download data from
data_origin_name (str) – Name of the downloaded b2 file
Examples
>>> get_bz2_data("data_dir", "kdda.t", "https://www.csie.ntu.edu.tw/~cjlin/libsvmtools/datasets/binary/kdda.t.bz2", "kdda.t.bz2")
-
mxnet.test_utils.
get_cifar10
()[source]¶ Downloads CIFAR10 dataset into a directory in the current directory with the name data, and then extracts all files into the directory data/cifar.
-
mxnet.test_utils.
get_im2rec_path
(home_env='MXNET_HOME')[source]¶ Get path to the im2rec.py tool
- Parameters
home_env (str) – Env variable that holds the path to the MXNET folder
- Returns
The path to im2rec.py
- Return type
str
-
mxnet.test_utils.
get_mnist
()[source]¶ Download and load the MNIST dataset
- Returns
A dict containing the data
- Return type
dict
-
mxnet.test_utils.
get_mnist_iterator
(batch_size, input_shape, num_parts=1, part_index=0)[source]¶ Returns training and validation iterators for MNIST dataset
-
mxnet.test_utils.
get_mnist_pkl
()[source]¶ Downloads MNIST dataset as a pkl.gz into a directory in the current directory with the name data
-
mxnet.test_utils.
get_mnist_ubyte
()[source]¶ Downloads ubyte version of the MNIST dataset into a directory in the current directory with the name data and extracts all files in the zip archive to this directory.
-
mxnet.test_utils.
get_zip_data
(data_dir, url, data_origin_name)[source]¶ Download and extract zip data.
- Parameters
data_dir (str) – Absolute or relative path of the directory name to store zip files
url (str) – URL to download data from
data_origin_name (str) – Name of the downloaded zip file
Examples
>>> get_zip_data("data_dir", "http://files.grouplens.org/datasets/movielens/ml-10m.zip", "ml-10m.zip")
-
mxnet.test_utils.
has_tvm_ops
()[source]¶ Returns True if MXNet is compiled with TVM generated operators. If current ctx is GPU, it only returns True for CUDA compute capability > 52 where FP16 is supported.
-
mxnet.test_utils.
is_cd_run
()[source]¶ Checks if the test is running as part of a Continuous Delivery run
-
mxnet.test_utils.
is_op_runnable
()[source]¶ Returns True for all CPU tests. Returns True for GPU tests that are either of the following. 1. Built with USE_TVM_OP=0. 2. Built with USE_TVM_OP=1, but with compute capability >= 53.
-
mxnet.test_utils.
list_gpus
()[source]¶ Return a list of GPUs
- Returns
If there are n GPUs, then return a list [0,1,…,n-1]. Otherwise returns [].
- Return type
list of int
-
mxnet.test_utils.
locationError
(a, b, index, names, maxError=False)[source]¶ Create element mismatch comment
- Parameters
b (a,) –
index (tuple of coordinate arrays) – Location of violation
names (tuple of names) – The names of compared arrays.
maxError (boolean, optional) – Flag indicating that maximum error is reporting.
-
mxnet.test_utils.
mean_check
(generator, mu, sigma, nsamples=1000000)[source]¶ Test the generator by matching the mean.
- We test the sample mean by checking if it falls inside the range
(mu - 3 * sigma / sqrt(n), mu + 3 * sigma / sqrt(n))
References:
@incollection{goucher2009beautiful, title={Beautiful Testing: Leading Professionals Reveal How They Improve Software}, author={Goucher, Adam and Riley, Tim}, year={2009}, chapter=10 }
Examples:
generator = lambda x: np.random.normal(0, 1.0, size=x) mean_check_ret = mean_check(generator, 0, 1.0)
- Parameters
generator (function) – The generator function. It’s expected to generate N i.i.d samples by calling generator(N).
mu (float) –
sigma (float) –
nsamples (int) –
- Returns
ret – Whether the mean test succeeds
- Return type
bool
-
mxnet.test_utils.
new_matrix_with_real_eigvals_2d
(n)[source]¶ Generate a well-conditioned matrix with small real eigenvalues.
-
mxnet.test_utils.
new_matrix_with_real_eigvals_nd
(shape)[source]¶ Generate well-conditioned matrices with small real eigenvalues.
-
mxnet.test_utils.
new_sym_matrix_with_real_eigvals_2d
(n)[source]¶ Generate a sym matrix with real eigenvalues.
-
mxnet.test_utils.
new_sym_matrix_with_real_eigvals_nd
(shape)[source]¶ Generate sym matrices with real eigenvalues.
-
mxnet.test_utils.
np_reduce
(dat, axis, keepdims, numpy_reduce_func)[source]¶ Compatible reduce for old version of NumPy.
- Parameters
dat (np.ndarray) – Same as NumPy.
axis (None or int or list-like) – Same as NumPy.
keepdims (bool) – Same as NumPy.
numpy_reduce_func (function) – A NumPy reducing function like
np.sum
ornp.max
.
-
mxnet.test_utils.
numeric_grad
(executor, location, aux_states=None, eps=0.0001, use_forward_train=True, dtype=<class 'numpy.float32'>)[source]¶ Calculates a numeric gradient via finite difference method.
Class based on Theano’s theano.gradient.numeric_grad [1]
- Parameters
executor (Executor) – Executor that computes the forward pass.
location (list of numpy.ndarray or dict of str to numpy.ndarray) – Argument values used as location to compute gradient Maps the name of arguments to the corresponding numpy.ndarray. Value of all the arguments must be provided.
aux_states (None or list of numpy.ndarray or dict of str to numpy.ndarray, optional) – Auxiliary states values used as location to compute gradient Maps the name of aux_states to the corresponding numpy.ndarray. Value of all the auxiliary arguments must be provided.
eps (float, optional) – Epsilon for the finite-difference method.
use_forward_train (bool, optional) – Whether to use is_train=True in testing.
dtype (np.float16 or np.float32 or np.float64) – Datatype for mx.nd.array.
References
..[1] https://github.com/Theano/Theano/blob/master/theano/gradient.py
-
mxnet.test_utils.
rand_ndarray
(shape, stype='default', density=None, dtype=None, modifier_func=None, shuffle_csr_indices=False, distribution=None, ctx=None)[source]¶ Generate a random sparse ndarray. Returns the generated ndarray.
-
mxnet.test_utils.
rand_sparse_ndarray
(shape, stype, density=None, dtype=None, distribution=None, data_init=None, rsp_indices=None, modifier_func=None, shuffle_csr_indices=False, ctx=None)[source]¶ Generate a random sparse ndarray. Returns the ndarray, value(np) and indices(np)
- Parameters
shape (list or tuple) –
stype (str) – valid values: “csr” or “row_sparse”
density (float, optional) – should be between 0 and 1
distribution (str, optional) – valid values: “uniform” or “powerlaw”
dtype (numpy.dtype, optional) – default value is None
- Returns
- Return type
Result of type CSRNDArray or RowSparseNDArray
Examples
Below is an example of the powerlaw distribution with csr as the stype. It calculates the nnz using the shape and density. It fills up the ndarray with exponentially increasing number of elements. If there are enough unused_nnzs, n+1th row will have twice more nnzs compared to nth row. else, remaining unused_nnzs will be used in n+1th row If number of cols is too small and we have already reached column size it will fill up all following columns in all followings rows until we reach the required density.
>>> csr_arr, _ = rand_sparse_ndarray(shape=(5, 16), stype="csr", density=0.50, distribution="powerlaw") >>> indptr = csr_arr.indptr.asnumpy() >>> indices = csr_arr.indices.asnumpy() >>> data = csr_arr.data.asnumpy() >>> row2nnz = len(data[indptr[1]:indptr[2]]) >>> row3nnz = len(data[indptr[2]:indptr[3]]) >>> assert(row3nnz == 2*row2nnz) >>> row4nnz = len(data[indptr[3]:indptr[4]]) >>> assert(row4nnz == 2*row3nnz)
-
mxnet.test_utils.
random_sample
(population, k)[source]¶ Return a k length list of the elements chosen from the population sequence.
-
mxnet.test_utils.
random_uniform_arrays
(*shapes, **kwargs)[source]¶ Generate some random numpy arrays.
-
mxnet.test_utils.
same
(a, b)[source]¶ Test if two NumPy arrays are the same.
- Parameters
a (np.ndarray) –
b (np.ndarray) –
-
mxnet.test_utils.
same_array
(array1, array2)[source]¶ Check whether two NDArrays sharing the same memory block
-
mxnet.test_utils.
same_symbol_structure
(sym1, sym2)[source]¶ Compare two symbols to check if they have the same computation graph structure. Returns true if operator corresponding to a particular node id is same in both symbols for all nodes
-
mxnet.test_utils.
set_env_var
(key, val, default_val='')[source]¶ Set environment variable
- Parameters
key (str) – Env var to set
val (str) – New value assigned to the env var
default_val (str, optional) – Default value returned if the env var doesn’t exist
- Returns
The value of env var before it is set to the new value
- Return type
str
-
mxnet.test_utils.
shuffle_csr_column_indices
(csr)[source]¶ Shuffle CSR column indices per row This allows validation of unordered column indices, which is not a requirement for a valid CSR matrix
-
mxnet.test_utils.
simple_forward
(sym, ctx=None, is_train=False, **inputs)[source]¶ A simple forward function for a symbol.
Primarily used in doctest to test the functionality of a symbol. Takes NumPy arrays as inputs and outputs are also converted to NumPy arrays.
- Parameters
ctx (Context) – If
None
, will take the default context.inputs (keyword arguments) – Mapping each input name to a NumPy array.
- Returns
The result as a numpy array. Multiple results will
be returned as a list of NumPy arrays.
-
mxnet.test_utils.
var_check
(generator, sigma, nsamples=1000000)[source]¶ Test the generator by matching the variance. It will need a large number of samples and is not recommended to use
- We test the sample variance by checking if it falls inside the range
(sigma^2 - 3 * sqrt(2 * sigma^4 / (n-1)), sigma^2 + 3 * sqrt(2 * sigma^4 / (n-1)))
References:
@incollection{goucher2009beautiful, title={Beautiful Testing: Leading Professionals Reveal How They Improve Software}, author={Goucher, Adam and Riley, Tim}, year={2009}, chapter=10 }
Examples:
generator = lambda x: np.random.normal(0, 1.0, size=x) var_check_ret = var_check(generator, 0, 1.0)
- Parameters
generator (function) – The generator function. It’s expected to generate N i.i.d samples by calling generator(N).
sigma (float) –
nsamples (int) –
- Returns
ret – Whether the variance test succeeds
- Return type
bool
-
mxnet.test_utils.
verify_generator
(generator, buckets, probs, nsamples=1000000, nrepeat=5, success_rate=0.2, alpha=0.05)[source]¶ Verify whether the generator is correct using chi-square testing.
- The test is repeated for “nrepeat” times and we check if the success rate is
above the threshold (25% by default).
- Parameters
generator (function) –
- A function that is assumed to generate i.i.d samples from a specific distribution.
generator(N) should generate N random samples.
buckets (list of tuple or list of number) –
- The buckets to run the chi-square the test. Make sure that the buckets cover
the whole range of the distribution. Also, the buckets must be in ascending order and have no intersection
probs (list or tuple) – The ground-truth probability of the random value fall in a specific bucket.
nsamples (int) – The number of samples to generate for the testing
nrepeat (int) – The times to repeat the test
success_rate (float) – The desired success rate
alpha (float) – The desired threshold for type-I error i.e. when a true null hypothesis is rejected
- Returns
cs_ret_l – The p values of the chi-square test.
- Return type
list
此页内容是否对您有帮助
感谢反馈!