-
Notifications
You must be signed in to change notification settings - Fork 301
PyTorch DirectML Operator Roadmap
Xiang Zhang edited this page Apr 27, 2023
·
6 revisions
The table below lists PyTorch operators that either have a GPU kernel implemented or a default backend implementation that has been verified to work in our tests.
Please note that this is not a comprehensive list of all PyTorch operators, and many operators have a default implementation that might not be listed here and still work. If you would require support for an operator that is not on this list, please file an issue.
| op_name | supported_data_types | |
|---|---|---|
| 0 | aten::abs | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 1 | aten::abs_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 2 | aten::acos | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 3 | aten::acos_ | [torch.float16, torch.float32] |
| 4 | aten::acosh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 5 | aten::acosh_ | [torch.float16, torch.float32] |
| 6 | aten::_adaptive_avg_pool2d | [torch.float16, torch.float32] |
| 7 | aten::adaptive_avg_pool2d | [torch.float16, torch.float32] |
| 8 | aten::_adaptive_avg_pool2d_backward | [torch.float16, torch.float32] |
| 9 | aten::add | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 10 | aten::add_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 11 | aten::addcdiv | [torch.float16, torch.float32, torch.float64] |
| 12 | aten::addcdiv_ | [torch.float16, torch.float32, torch.float64] |
| 13 | aten::addcmul | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 14 | aten::addcmul_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 15 | aten::addmm | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 16 | aten::alias | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 17 | aten::all | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 18 | aten::__and__ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 19 | aten::any | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 20 | aten::arange | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 21 | aten::argmax | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 22 | aten::argmin | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 23 | aten::asin | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 24 | aten::asin_ | [torch.float16, torch.float32] |
| 25 | aten::asinh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 26 | aten::asinh_ | [torch.float16, torch.float32] |
| 27 | aten::as_strided | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 28 | aten::as_strided_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 29 | aten::atan | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 30 | aten::atan_ | [torch.float16, torch.float32] |
| 31 | aten::atan2 | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 32 | aten::atan2_ | [torch.float16, torch.float32] |
| 33 | aten::atanh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 34 | aten::atanh_ | [torch.float16, torch.float32] |
| 35 | aten::avg_pool2d | [torch.float16, torch.float32] |
| 36 | aten::avg_pool2d_backward | [torch.float16, torch.float32] |
| 37 | aten::baddbmm | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 38 | aten::batch_norm | [torch.float32] |
| 39 | aten::_batch_norm_impl_index | [torch.float32] |
| 40 | aten::bernoulli_ | [torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 41 | aten::bernoulli | [torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 42 | aten::binary_cross_entropy_with_logits | [torch.float32] |
| 43 | aten::bitwise_and | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 44 | aten::bitwise_and_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 45 | aten::bitwise_not | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 46 | aten::bitwise_not_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 47 | aten::bitwise_or | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 48 | aten::bitwise_or_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 49 | aten::bitwise_xor | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 50 | aten::bitwise_xor_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 51 | aten::bmm | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 52 | aten::broadcast_tensors | [torch.float16, torch.float32] |
| 53 | aten::broadcast_to | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 54 | aten::can_cast | [torch.float32, torch.float16] |
| 55 | aten::cat | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 56 | aten::cdist | [torch.float32] |
| 57 | aten::_cdist_forward | [torch.float32] |
| 58 | aten::ceil | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64] |
| 59 | aten::ceil_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64] |
| 60 | aten::clamp | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 61 | aten::clamp_ | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 62 | aten::clamp_max | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16] |
| 63 | aten::clamp_max_ | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16] |
| 64 | aten::clamp_min | [torch.float32, torch.float16, torch.int32, torch.int64, torch.int16] |
| 65 | aten::clamp_min_ | [torch.float32, torch.float16, torch.int32, torch.int64, torch.int16] |
| 66 | aten::clone | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 67 | aten::constant_pad_nd | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 68 | aten::conv2d | [torch.float32] |
| 69 | aten::_convolution | [torch.float32] |
| 70 | aten::convolution | [torch.float32] |
| 71 | aten::convolution_backward | [torch.float32] |
| 72 | aten::convolution_backward_overrideable | [torch.float32] |
| 73 | aten::convolution_overrideable | [torch.float32] |
| 74 | aten::conv_transpose2d | [torch.float32] |
| 75 | aten::copy_ | [torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16] |
| 76 | aten::cos | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 77 | aten::cos_ | [torch.float16, torch.float32] |
| 78 | aten::cosh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 79 | aten::cosh_ | [torch.float16, torch.float32] |
| 80 | aten::cudnn_is_acceptable | [torch.float32] |
| 81 | aten::cumsum | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 82 | aten::detach | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 83 | aten::diag | [torch.float16, torch.float32] |
| 84 | aten::diagonal | [torch.float16, torch.float32] |
| 85 | aten::diagonal_backward | [torch.float16, torch.float32] |
| 86 | aten::diagonal_copy | [torch.float16, torch.float32] |
| 87 | aten::div | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 88 | aten::div_ | [torch.float32, torch.float16, torch.float64] |
| 89 | aten::dot | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 90 | aten::dropout | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 91 | aten::dropout_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 92 | aten::_efficientzerotensor | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 93 | aten::embedding | [torch.int32, torch.int64, torch.float16, torch.float32, torch.float64] |
| 94 | aten::embedding_backward | [torch.float16, torch.float32, torch.float64] |
| 95 | aten::embedding_dense_backward | [torch.float16, torch.float32, torch.float64] |
| 96 | aten::empty | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16, torch.float32, torch.float64] |
| 97 | aten::empty_like | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 98 | aten::empty_strided | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 99 | aten::eq | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 100 | aten::eq_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 101 | aten::equal | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 102 | aten::erf | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 103 | aten::erf_ | [torch.float16, torch.float32, torch.float64] |
| 104 | aten::erfc | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 105 | aten::erfc_ | [torch.float16, torch.float32] |
| 106 | aten::exp | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 107 | aten::exp_ | [torch.float32, torch.float16, torch.float64] |
| 108 | aten::exp2 | [torch.float16, torch.float32] |
| 109 | aten::exp2_ | [torch.float16, torch.float32] |
| 110 | aten::expand | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 111 | aten::expand_as | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 112 | aten::expm1 | [torch.float16, torch.float32, torch.float64] |
| 113 | aten::expm1_ | [torch.float16, torch.float32, torch.float64] |
| 114 | aten::eye | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 115 | aten::fill_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 116 | aten::flatten | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 117 | aten::flip | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 118 | aten::floor | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64] |
| 119 | aten::floor_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64] |
| 120 | aten::floor_divide | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 121 | aten::floor_divide_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 122 | aten::frobenius_norm | [torch.float16, torch.float32] |
| 123 | aten::full | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 124 | aten::full_like | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 125 | aten::gather | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 126 | aten::ge | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 127 | aten::ge_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 128 | aten::gelu | [torch.float16, torch.float32] |
| 129 | aten::gelu_backward | [torch.float16, torch.float32] |
| 130 | aten::group_norm | [torch.float16, torch.float32] |
| 131 | aten::gru | [torch.float32] |
| 132 | aten::gt | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 133 | aten::gt_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 134 | aten::hardsigmoid | [torch.float16, torch.float32] |
| 135 | aten::hardsigmoid_ | [torch.float16, torch.float32] |
| 136 | aten::hardsigmoid_backward | [torch.float16, torch.float32] |
| 137 | aten::hardswish | [torch.float16, torch.float32] |
| 138 | aten::hardswish_ | [torch.float16, torch.float32] |
| 139 | aten::hardswish_backward | [torch.float16, torch.float32] |
| 140 | aten::hardtanh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 141 | aten::hardtanh_ | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 142 | aten::hardtanh_backward | [torch.float16, torch.float32] |
| 143 | aten::index_put_ | [torch.float32, torch.float64, torch.float16] |
| 144 | aten::index_put | [torch.float16, torch.float32, torch.float64] |
| 145 | aten::_index_put_impl_ | [torch.float32, torch.float64, torch.float16] |
| 146 | aten::index_select | [torch.int32, torch.int64, torch.float16, torch.float32, torch.float64, torch.int16, torch.int8, torch.bool] |
| 147 | aten::isfinite | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 148 | aten::isinf | [torch.float16, torch.float32] |
| 149 | aten::isnan | [torch.float16, torch.float32] |
| 150 | aten::is_nonzero | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 151 | aten::is_same_size | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 152 | aten::item | [torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16, torch.float64] |
| 153 | aten::layer_norm | [torch.float16, torch.float32] |
| 154 | aten::le | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 155 | aten::le_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 156 | aten::leaky_relu | [torch.float16, torch.float32] |
| 157 | aten::leaky_relu_backward | [torch.float16, torch.float32] |
| 158 | aten::less_equal | [torch.float32] |
| 159 | aten::linear | [torch.float32] |
| 160 | aten::_local_scalar_dense | [torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16, torch.float64] |
| 161 | aten::log | [torch.float32, torch.float16, torch.int32, torch.int64, torch.int16, torch.int8] |
| 162 | aten::log_ | [torch.float32, torch.float16] |
| 163 | aten::log10 | [torch.float16, torch.float32] |
| 164 | aten::log10_ | [torch.float16, torch.float32] |
| 165 | aten::log1p | [torch.float16, torch.float32] |
| 166 | aten::log1p_ | [torch.float16, torch.float32] |
| 167 | aten::log2 | [torch.float16, torch.float32] |
| 168 | aten::log2_ | [torch.float16, torch.float32] |
| 169 | aten::logical_and | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 170 | aten::logical_and_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 171 | aten::logical_not | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 172 | aten::logical_not_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 173 | aten::logical_or | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 174 | aten::logical_or_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 175 | aten::logical_xor | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 176 | aten::logical_xor_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 177 | aten::_log_softmax | [torch.float16, torch.float32] |
| 178 | aten::log_softmax | [torch.float16, torch.float32] |
| 179 | aten::_log_softmax_backward_data | [torch.float16, torch.float32] |
| 180 | aten::lt | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 181 | aten::lt_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 182 | aten::masked_fill_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 183 | aten::masked_scatter_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 184 | aten::masked_select | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 185 | aten::matmul | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 186 | aten::max | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 187 | aten::maximum | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 188 | aten::max_pool2d | [torch.float16, torch.float32] |
| 189 | aten::max_pool2d_with_indices | [torch.float16, torch.float32] |
| 190 | aten::max_pool2d_with_indices_backward | [torch.float16, torch.float32] |
| 191 | aten::mean | [torch.float16, torch.float32, torch.float64] |
| 192 | aten::meshgrid | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 193 | aten::min | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 194 | aten::minimum | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 195 | aten::mm | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 196 | aten::mse_loss | [torch.float16, torch.float32] |
| 197 | aten::mul | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 198 | aten::mul_ | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 199 | aten::narrow | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 200 | aten::native_batch_norm | [torch.float32, torch.float16] |
| 201 | aten::native_batch_norm_backward | [torch.float32] |
| 202 | aten::native_group_norm | [torch.float16, torch.float32] |
| 203 | aten::native_layer_norm | [torch.float16, torch.float32] |
| 204 | aten::native_layer_norm_backward | [torch.float32] |
| 205 | aten::ne | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 206 | aten::ne_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 207 | aten::neg | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 208 | aten::neg_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 209 | aten::new_empty | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 210 | aten::new_full | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 211 | aten::new_ones | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 212 | aten::new_zeros | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 213 | aten::nll_loss | [torch.float32] |
| 214 | aten::nll_loss_forward | [torch.float32] |
| 215 | aten::nll_loss_nd | [torch.float32] |
| 216 | aten::nonzero | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 217 | aten::norm | [torch.float16, torch.float32, torch.float64] |
| 218 | aten::normal_ | [torch.float16, torch.float32, torch.float64] |
| 219 | aten::ones | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 220 | aten::ones_like | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 221 | aten::__or__ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 222 | aten::pad | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 223 | aten::permute | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 224 | aten::pow | [torch.float32, torch.float16, torch.bool] |
| 225 | aten::pow_ | [torch.float32, torch.float16, torch.bool] |
| 226 | aten::prelu | [torch.float32] |
| 227 | aten::_prelu_kernel | [torch.float32] |
| 228 | aten::_prelu_kernel_backward | [torch.float32] |
| 229 | aten::prod | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 230 | aten::random_ | [torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 231 | aten::randperm | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 232 | aten::reciprocal | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 233 | aten::reciprocal_ | [torch.float16, torch.float32] |
| 234 | aten::reflection_pad1d | [torch.float16, torch.float32, torch.float64] |
| 235 | aten::reflection_pad1d_backward | [torch.float16, torch.float32] |
| 236 | aten::reflection_pad2d | [torch.float16, torch.float32, torch.float64] |
| 237 | aten::reflection_pad2d_backward | [torch.float16, torch.float32] |
| 238 | aten::relu | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 239 | aten::relu_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 240 | aten::remainder | [torch.int32, torch.int64, torch.int16, torch.int8] |
| 241 | aten::remainder_ | [torch.int32, torch.int64, torch.int16, torch.int8] |
| 242 | aten::repeat | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 243 | aten::replication_pad1d | [torch.float16, torch.float32, torch.float64] |
| 244 | aten::replication_pad2d | [torch.float16, torch.float32, torch.float64] |
| 245 | aten::reshape | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 246 | aten::_reshape_alias | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 247 | aten::reshape_as | [torch.float32] |
| 248 | aten::resize_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 249 | aten::resize_as_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 250 | aten::result_type | [torch.float32, torch.float16] |
| 251 | aten::round | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 252 | aten::round_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 253 | aten::rsqrt | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 254 | aten::rsqrt_ | [torch.float16, torch.float32] |
| 255 | aten::rsub | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 256 | aten::scalar_tensor | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 257 | aten::scatter_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 258 | aten::scatter | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 259 | aten::select | [torch.float32] |
| 260 | aten::sgn | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 261 | aten::sigmoid | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 262 | aten::sigmoid_ | [torch.float32, torch.float16, torch.float64] |
| 263 | aten::sigmoid_backward | [torch.float32] |
| 264 | aten::sign | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 265 | aten::silu | [torch.float16, torch.float32] |
| 266 | aten::silu_ | [torch.float16, torch.float32] |
| 267 | aten::sin | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 268 | aten::sin_ | [torch.float16, torch.float32] |
| 269 | aten::sinc | [torch.float16, torch.float32] |
| 270 | aten::sinc_ | [torch.float16, torch.float32] |
| 271 | aten::sinh | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 272 | aten::sinh_ | [torch.float16, torch.float32] |
| 273 | aten::slice | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 274 | aten::slice_backward | [torch.float16, torch.float32, torch.float64] |
| 275 | aten::smooth_l1_loss | [torch.float16, torch.float32] |
| 276 | aten::softmax | [torch.float16, torch.float32] |
| 277 | aten::_softmax | [torch.float16, torch.float32] |
| 278 | aten::_softmax_backward_data | [torch.float16, torch.float32] |
| 279 | aten::sort | [torch.float16, torch.float32, torch.int32, torch.int16, torch.int8, torch.bool] |
| 280 | aten::split | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 281 | aten::split_with_sizes | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 282 | aten::sqrt | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 283 | aten::sqrt_ | [torch.float16, torch.float32] |
| 284 | aten::squeeze | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 285 | aten::stack | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 286 | aten::sub | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 287 | aten::sub_ | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 288 | aten::sum | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8] |
| 289 | aten::t | [torch.float32, torch.float16, torch.float64] |
| 290 | aten::tan | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 291 | aten::tan_ | [torch.float16, torch.float32] |
| 292 | aten::tanh | [torch.float32, torch.float16, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 293 | aten::tanh_ | [torch.float32, torch.float16] |
| 294 | aten::tanh_backward | [torch.float32] |
| 295 | aten::threshold | [torch.float32] |
| 296 | aten::threshold_backward | [torch.float16, torch.float32] |
| 297 | aten::to | [torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16] |
| 298 | aten::_to_copy | [torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool, torch.float16] |
| 299 | aten::topk | [torch.float16, torch.float32, torch.int32, torch.int64, torch.int16, torch.int8] |
| 300 | aten::transpose | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 301 | aten::transpose_ | [torch.float32] |
| 302 | aten::triu | [torch.float16, torch.float32, torch.float64, torch.int32] |
| 303 | aten::type_as | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 304 | aten::unbind | [torch.float32, torch.float16, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 305 | aten::unfold | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 306 | aten::_unique2 | [torch.float32, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 307 | aten::unsafe_chunk | [torch.float32] |
| 308 | aten::unsafe_split | [torch.float32] |
| 309 | aten::unsqueeze_ | [torch.float32] |
| 310 | aten::unsqueeze | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 311 | aten::upsample_bilinear2d | [torch.float16, torch.float32] |
| 312 | aten::upsample_bilinear2d_backward | [torch.float16, torch.float32] |
| 313 | aten::upsample_nearest2d | [torch.float16, torch.float32] |
| 314 | aten::upsample_nearest2d_backward | [torch.float16, torch.float32] |
| 315 | aten::var | [torch.float16, torch.float32] |
| 316 | aten::view | [torch.int32, torch.int64, torch.float16, torch.float32, torch.float64, torch.int16, torch.int8, torch.bool] |
| 317 | aten::where | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 318 | aten::zero_ | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |
| 319 | aten::zeros | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int16, torch.int8, torch.bool, torch.int64] |
| 320 | aten::zeros_like | [torch.float16, torch.float32, torch.float64, torch.int32, torch.int64, torch.int16, torch.int8, torch.bool] |