Thanks for sending an issue! Please fill in the following template to help quickly solve your problem.
quant/mx_to_block_mx_quant/docs/torchapi_mx_to_block_mx_quant.md
y, scale1, scale2 = torch.ops.cann_ops_nn.mx_to_block_mx_quant( x, mxscale, dst_type=torch_npu.float8_e5m2, x_type=torch_npu.float4_e1m2fn_x2 ) return torch.ops.cann_ops_nn.mx_to_block_mx_quant( x, mxscale, dst_type=dst_type, x_type=x_type )
整改调用方式
/assign
Thanks for sending an issue! Please fill in the following template to help quickly solve your problem.
Document Link(文档链接)
quant/mx_to_block_mx_quant/docs/torchapi_mx_to_block_mx_quant.md
Issues Section(问题文档片段)
Existing Issues(存在的问题)
整改调用方式