export_matrix.csv

operator,forward_lowering_min_opset,gradient_lowering_min_opset
:op/and,1,
:op/idx_arr,1,
:op/idx_take,1,
:op/not,1,
:op/or,1,
:op/slice_sub,10,
:tangent/scatter_slice_arr,13,
autodiff:stop_gradient,,1
math:sqrt,1,
nn:adam,,7
nn:adam_state,,8
nn:avg_pool2d,1,
nn:avg_pool2d_grad,,
nn:avg_pool_f32,,
nn:batch_norm,15,
nn:batch_norm_train,13,
nn:batch_norm_train_grad,,13
nn:conv2d,1,
nn:conv2d_bias_grad,,1
nn:conv2d_input_grad,,1
nn:conv2d_kernel_grad,,10
nn:embedding,1,
nn:embedding_table_grad,,
nn:max_pool2d,1,
nn:max_pool2d_grad,,11
nn:max_pool_f32,,
nn:sgd,,1
nn:softmax_cross_entropy,,
nn:softmax_cross_entropy_grad,,
nn:softmax_cross_entropy_labels,,
nn:softmax_cross_entropy_labels_grad,,
tensor:abs,1,
tensor:add,1,
tensor:argmax_axis,12,
tensor:broadcast_like,,8
tensor:cast_like,,1
tensor:concat,1,
tensor:dim,1,
tensor:divide,1,
tensor:erf,1,
tensor:erf_grad,,1
tensor:exp,1,
tensor:expand_axis,,13
tensor:eye,,
tensor:fma_f32,,
tensor:from_bytes_aggregate,,
tensor:full,9,
tensor:ge,1,
tensor:gelu,1,
tensor:gelu_grad,,1
tensor:gemm_backend,,
tensor:gemm_library,,
tensor:gemm_library_threads,,
tensor:gemm_threads,,
tensor:gt,1,
tensor:idx,1,
tensor:idx2d,1,
tensor:idx_take,,
tensor:layer_norm,17,
tensor:layer_norm_grad,,13
tensor:le,1,
tensor:linear,1,
tensor:log,1,
tensor:log_softmax,13,
tensor:lt,1,
tensor:matmul,1,
tensor:matmul_grad_lhs,,13
tensor:matmul_grad_rhs,,13
tensor:max_axis,1,
tensor:maximum,1,
tensor:mean,1,
tensor:mean_axis,1,
tensor:min_axis,1,
tensor:minimum,1,
tensor:mul_add_i64,,
tensor:multiply,1,
tensor:neg,1,
tensor:num_threads,,
tensor:numel,1,
tensor:one_hot,,
tensor:ones,9,
tensor:ones_like,,1
tensor:permute,1,
tensor:permute_inverse,,1
tensor:pow,1,
tensor:relu,1,
tensor:reshape,5,
tensor:row_add_at,,
tensor:rsqrt,1,
tensor:save_npy,,
tensor:set_num_threads,,
tensor:shape,1,
tensor:show,,
tensor:sigmoid,1,
tensor:softmax,13,
tensor:softmax_grad,,13
tensor:sqrt,1,
tensor:strict_eq,1,
tensor:strict_neq,1,
tensor:subtract,1,
tensor:sum,1,
tensor:sum_axis,1,
tensor:sum_to,,13
tensor:take,1,
tensor:take_grad,,
tensor:tanh,1,
tensor:to_bytes,,
tensor:transpose,1,
tensor:unsqueeze,13,
tensor:where,1,
tensor:zeros,9,
tensor:zeros_like,,1
tiles:tiles_,,
tiles:tiles_rows,,