接口支持清单

算子名	Torch API	ascend910b	ascend910_93	ascend950
abs	torch.ops.ops_multimodal_fusion.abs(Tensor x) -> Tensor	✓	✓	-
upsample_linear1d	torch.ops.ops_multimodal_fusion.upsample_linear1d(Tensor input, int output_size, bool align_corners=False, float scale=-1.) -> Tensor	-	-	✓
adaptive_avg_pool2d	torch.ops.ops_multimodal_fusion.adaptive_avg_pool2d(Tensor self, int[2] output_size) -> Tensor	-	-	✓
angle	torch.ops.ops_multimodal_fusion.angle(Tensor input) -> Tensor	-	-	✓
bessel_j0	torch.ops.ops_multimodal_fusion.bessel_j0(Tensor x) -> Tensor	-	-	✓
bessel_j1	torch.ops.ops_multimodal_fusion.bessel_j1(Tensor x) -> Tensor	-	-	✓
bessel_y0	torch.ops.ops_multimodal_fusion.bessel_y0(Tensor x) -> Tensor	-	-	✓
bessel_y1	torch.ops.ops_multimodal_fusion.bessel_y1(Tensor x) -> Tensor	-	-	✓
cauchy	torch.ops.ops_multimodal_fusion.cauchy(Tensor x, float median=0.0, float sigma=1.0, int seed=0) -> Tensor	-	-	✓
complex	torch.ops.ops_multimodal_fusion.complex(Tensor real, Tensor imag) -> Tensor	-	-	✓
conjphysical	torch.ops.ops_multimodal_fusion.conjphysical(Tensor x) -> Tensor	-	-	✓
cummax	torch.ops.ops_multimodal_fusion.cummax(Tensor x, int dim) -> (Tensor values, Tensor indices)	-	-	✓
cumprod	torch.ops.ops_multimodal_fusion.cumprod(Tensor x, int dim) -> Tensor	-	-	✓
depthwise_conv3d	torch.ops.ops_multimodal_fusion.depthwise_conv3d(Tensor input, Tensor weight, Tensor? bias, int[3] stride, int[3] padding, int[3] dilation) -> Tensor	-	-	✓
digamma	torch.ops.ops_multimodal_fusion.digamma(Tensor x) -> Tensor	-	-	✓
entr	torch.ops.ops_multimodal_fusion.entr(Tensor x) -> Tensor	-	-	✓
erfcx	torch.ops.ops_multimodal_fusion.erfcx(Tensor x) -> Tensor	-	-	✓
foreach_ceil	torch.ops.ops_multimodal_fusion.foreach_ceil(Tensor[] tensors) -> Tensor[]	-	-	✓
index_reduce	torch.ops.ops_multimodal_fusion.index_reduce(Tensor self, int dim, Tensor index, Tensor source, str reduce, bool include_self=True) -> Tensor	-	-	✓
int_repr	torch.ops.ops_multimodal_fusion.int_repr(Tensor x) -> Tensor	-	-	✓
kthvalue	torch.ops.ops_multimodal_fusion.kthvalue(Tensor x, int k, int dim, bool keepdim) -> (Tensor values, Tensor indices)	-	-	✓
log_normal	torch.ops.ops_multimodal_fusion.log_normal(Tensor x, float mean=1.0, float std=2.0, int seed=0) -> Tensor	-	-	✓
logndtr	torch.ops.ops_multimodal_fusion.logndtr(Tensor x) -> Tensor	-	-	✓
make_per_tensor_quantized	torch.ops.ops_multimodal_fusion.make_per_tensor_quantized(Tensor x, float scale, int zero_point) -> Tensor	-	-	✓
mode	torch.ops.ops_multimodal_fusion.mode(Tensor x, int dim, bool keepdim) -> (Tensor values, Tensor indices)	-	-	✓
multi_margin_loss	torch.ops.ops_multimodal_fusion.multi_margin_loss(Tensor input, Tensor target, Scalar p=1, Scalar margin=1.0, Tensor? weight=None, int reduction=1) -> Tensor	-	-	✓
multilabel_margin_loss	torch.ops.ops_multimodal_fusion.multilabel_margin_loss(Tensor input, Tensor target, int reduction=1) -> Tensor	-	-	✓
polygamma	torch.ops.ops_multimodal_fusion.polygamma(Tensor x, int n) -> Tensor	-	-	✓
searchsorted	torch.ops.ops_multimodal_fusion.searchsorted(Tensor sorted_sequence, Tensor values, bool out_int32=False, bool right=False) -> Tensor	-	-	✓
shifted_chebyshev_polynomial_t	torch.ops.ops_multimodal_fusion.shifted_chebyshev_polynomial_t(Tensor x, Tensor n) -> Tensor	-	-	✓
shifted_chebyshev_polynomial_u	torch.ops.ops_multimodal_fusion.shifted_chebyshev_polynomial_u(Tensor x, Tensor n) -> Tensor	-	-	✓
shifted_chebyshev_polynomial_v	torch.ops.ops_multimodal_fusion.shifted_chebyshev_polynomial_v(Tensor x, Tensor n) -> Tensor	-	-	✓
shifted_chebyshev_polynomial_w	torch.ops.ops_multimodal_fusion.shifted_chebyshev_polynomial_w(Tensor x, Tensor n) -> Tensor	-	-	✓
tril_indices	torch.ops.ops_multimodal_fusion.tril_indices(int row, int col, int offset, bool out_int32) -> Tensor	-	-	✓
triu_indices	torch.ops.ops_multimodal_fusion.triu_indices(int row, int col, int offset, bool out_int32) -> Tensor	-	-	✓
upsample_trilinear3d	torch.ops.ops_multimodal_fusion.upsample_trilinear3d(Tensor input, int[3] output_size, bool align_corners=False, float scales_d=-1., float scales_h=-1., float scales_w=-1.) -> Tensor	-	-	✓

✓ 表示已支持，- 表示暂未支持。详细的接口说明请参考接口列表。