ONNX supported ATen operators

This file is automatically generated during the documentation build by cross referencing ONNX operator symbolics with Torch JIT operators via docs/source/scripts/build_onnx_supported_aten_op_csv_table.py. Do not modify directly and instead rebuild the docs.

Supported ATen operators

Operator	opset_version(s)
Delete	11 12 13 14 15 16
__derive_index	7 8 9 10 11 12 13 14 15 16
__interpolate	7 8 9 10 11 12 13 14 15 16
__range_length	7 8 9 10 11 12 13 14 15 16
_cast_Byte	7 8 9 10 11 12 13 14 15 16
_cast_Char	7 8 9 10 11 12 13 14 15 16
_cast_Double	7 8 9 10 11 12 13 14 15 16
_cast_Float	7 8 9 10 11 12 13 14 15 16
_cast_Half	7 8 9 10 11 12 13 14 15 16
_cast_Int	7 8 9 10 11 12 13 14 15 16
_cast_Long	7 8 9 10 11 12 13 14 15 16
_cast_Short	7 8 9 10 11 12 13 14 15 16
_convolution	7 8 9 10 11 12 13 14 15 16
_dim_arange	7 8 9 10 11 12 13 14 15 16
_pack_padded_sequence	7 8 9 10 11 12 13 14 15 16
_pad_circular	7 8 9 10 11 12 13 14 15 16
_pad_packed_sequence	7 8 9 10 11 12 13 14 15 16
_reshape_from_tensor	7 8 9 10 11 12 13 14 15 16
_sample_dirichlet	7 8 9 10 11 12 13 14 15 16
_set_item	11 12 13 14 15 16
_shape_as_tensor	7 8 9 10 11 12 13 14 15 16
_standard_gamma	7 8 9 10 11 12 13 14 15 16
_unique	7 8 9 10 11 12 13 14 15 16
_unique2	7 8 9 10 11 12 13 14 15 16
_weight_norm	7 8 9 10 11 12 13 14 15 16
abs	7 8 9 10 11 12 13 14 15 16
acos	7 8 9 10 11 12 13 14 15 16
adaptive_avg_pool1d	7 8 9 10 11 12 13 14 15 16
adaptive_avg_pool2d	7 8 9 10 11 12 13 14 15 16
adaptive_avg_pool3d	7 8 9 10 11 12 13 14 15 16
adaptive_max_pool1d	7 8 9 10 11 12 13 14 15 16
adaptive_max_pool2d	7 8 9 10 11 12 13 14 15 16
adaptive_max_pool3d	7 8 9 10 11 12 13 14 15 16
add	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
addcmul	7 8 9 10 11 12 13 14 15 16
addmm	7 8 9 10 11 12 13 14 15 16
alias	7 8 9 10 11 12 13 14 15 16
all	7 8 9 10 11 12 13 14 15 16
alpha_dropout	7 8 9 10 11 12 13 14 15 16
alpha_dropout_	7 8 9 10 11 12 13 14 15 16
amax	7 8 9 10 11 12 13 14 15 16
amin	7 8 9 10 11 12 13 14 15 16
aminmax	7 8 9 10 11 12 13 14 15 16
any	7 8 9 10 11 12 13 14 15 16
append	11 12 13 14 15 16
arange	7 8 9 10 11 12 13 14 15 16
argmax	7 8 9 10 11 12 13 14 15 16
argmin	7 8 9 10 11 12 13 14 15 16
as_strided	7 8 9 10 11 12 13 14 15 16
as_tensor	7 8 9 10 11 12 13 14 15 16
asin	7 8 9 10 11 12 13 14 15 16
atan	7 8 9 10 11 12 13 14 15 16
avg_pool1d	7 8 9 10 11 12 13 14 15 16
avg_pool2d	7 8 9 10 11 12 13 14 15 16
avg_pool3d	7 8 9 10 11 12 13 14 15 16
baddbmm	7 8 9 10 11 12 13 14 15 16
batch_norm	7 8 9 10 11 12 13 14 15 16
bernoulli	7 8 9 10 11 12 13 14 15 16
binary_cross_entropy_with_logits	12 13 14 15 16
bitwise_not	7 8 9 10 11 12 13 14 15 16
bmm	7 8 9 10 11 12 13 14 15 16
broadcast_tensors	7 8 9 10 11 12 13 14 15 16
bucketize	7 8 9 10 11 12 13 14 15 16
cat	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
cdist	7 8 9 10 11 12 13 14 15 16
ceil	7 8 9 10 11 12 13 14 15 16
celu	12 13 14 15 16
chunk	11 12 13 14 15 16
clamp	7 8 9 10 11 12 13 14 15 16
clamp_max	7 8 9 10 11 12 13 14 15 16
clamp_min	7 8 9 10 11 12 13 14 15 16
clone	7 8 9 10 11 12 13 14 15 16
constant_pad_nd	7 8 9 10 11 12 13 14 15 16
contiguous	7 8 9 10 11 12 13 14 15 16
conv1d	7 8 9 10 11 12 13 14 15 16
conv2d	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
conv3d	7 8 9 10 11 12 13 14 15 16
conv_tbc	7 8 9 10 11 12 13 14 15 16
conv_transpose1d	7 8 9 10 11 12 13 14 15 16
conv_transpose2d	7 8 9 10 11 12 13 14 15 16
conv_transpose3d	7 8 9 10 11 12 13 14 15 16
cos	7 8 9 10 11 12 13 14 15 16
cosine_similarity	7 8 9 10 11 12 13 14 15 16
cross	7 8 9 10 11 12 13 14 15 16
cross_entropy_loss	12 13 14 15 16
cumsum	7 8 9 10 11 12 13 14 15 16
data	7 8 9 10 11 12 13 14 15 16
dequantize	10 11 12 13 14 15 16
detach	7 8 9 10 11 12 13 14 15 16
device	7 8 9 10 11 12 13 14 15 16
diagonal	13 14 15 16
dim	7 8 9 10 11 12 13 14 15 16
div	7 8 9 10 11 12 13 14 15 16
dot	7 8 9 10 11 12 13 14 15 16
dropout	7 8 9 10 11 12 13 14 15 16
dropout_	7 8 9 10 11 12 13 14 15 16
einsum	12 13 14 15 16
elu	7 8 9 10 11 12 13 14 15 16
embedding	7 8 9 10 11 12 13 14 15 16
embedding_bag	7 8 9 10 11 12 13 14 15 16
embedding_renorm	11 12 13 14 15 16
empty	7 8 9 10 11 12 13 14 15 16
empty_like	7 8 9 10 11 12 13 14 15 16
eq	7 8 9 10 11 12 13 14 15 16
erf	7 8 9 10 11 12 13 14 15 16
exp	7 8 9 10 11 12 13 14 15 16
expand	7 8 9 10 11 12 13 14 15 16
expand_as	7 8 9 10 11 12 13 14 15 16
eye	7 8 9 10 11 12 13 14 15 16
fake_quantize_per_channel_affine	13 14 15 16
fake_quantize_per_tensor_affine	10 11 12 13 14 15 16
feature_alpha_dropout	7 8 9 10 11 12 13 14 15 16
feature_alpha_dropout_	7 8 9 10 11 12 13 14 15 16
feature_dropout	7 8 9 10 11 12 13 14 15 16
feature_dropout_	7 8 9 10 11 12 13 14 15 16
fill	7 8 9 10 11 12 13 14 15 16
flatten	7 8 9 10 11 12 13 14 15 16
flip	10 11 12 13 14 15 16
floor	7 8 9 10 11 12 13 14 15 16
floor_divide	7 8 9 10 11 12 13 14 15 16
floordiv	7 8 9 10 11 12 13 14 15 16
fmod	10 11 12 13 14 15 16
frobenius_norm	7 8 9 10 11 12 13 14 15 16
full	7 8 9 10 11 12 13 14 15 16
full_like	7 8 9 10 11 12 13 14 15 16
gather	7 8 9 10 11 12 13 14 15 16
ge	7 8 9 10 11 12 13 14 15 16
gelu	7 8 9 10 11 12 13 14 15 16
glu	7 8 9 10 11 12 13 14 15 16
grid_sampler	16
group_norm	7 8 9 10 11 12 13 14 15 16
gru	7 8 9 10 11 12 13 14 15 16
gt	7 8 9 10 11 12 13 14 15 16
hann_window	7 8 9 10 11 12 13 14 15 16
hardshrink	7 8 9 10 11 12 13 14 15 16
hardsigmoid	7 8 9 10 11 12 13 14 15 16
hardswish	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
hardtanh	7 8 9 10 11 12 13 14 15 16
im2col	11 12 13 14 15 16
index	7 8 9 10 11 12 13 14 15 16
index_add	7 8 9 10 11 12 13 14 15 16
index_copy	7 8 9 10 11 12 13 14 15 16
index_fill	7 8 9 10 11 12 13 14 15 16
index_put	7 8 9 10 11 12 13 14 15 16
index_select	7 8 9 10 11 12 13 14 15 16
insert	11 12 13 14 15 16
instance_norm	7 8 9 10 11 12 13 14 15 16
is_floating_point	7 8 9 10 11 12 13 14 15 16
isfinite	10 11 12 13 14 15 16
isinf	10 11 12 13 14 15 16
isnan	7 8 9 10 11 12 13 14 15 16
item	7 8 9 10 11 12 13 14 15 16
kl_div	7 8 9 10 11 12 13 14 15 16
layer_norm	7 8 9 10 11 12 13 14 15 16
le	7 8 9 10 11 12 13 14 15 16
leaky_relu	7 8 9 10 11 12 13 14 15 16
len	7 8 9 10 11 12 13 14 15 16
lift	7 8 9 10 11 12 13 14 15 16
linalg_cross	7 8 9 10 11 12 13 14 15 16
linalg_det	11 12 13 14 15 16
linalg_matrix_norm	7 8 9 10 11 12 13 14 15 16
linalg_norm	7 8 9 10 11 12 13 14 15 16
linalg_vector_norm	7 8 9 10 11 12 13 14 15 16
linear	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
linspace	7 8 9 10 11 12 13 14 15 16
list	7 8 9 10 11 12 13 14 15 16
log	7 8 9 10 11 12 13 14 15 16
log10	7 8 9 10 11 12 13 14 15 16
log1p	7 8 9 10 11 12 13 14 15 16
log2	7 8 9 10 11 12 13 14 15 16
log_sigmoid	7 8 9 10 11 12 13 14 15 16
log_softmax	7 8 9 10 11 12 13 14 15 16
logdet	11 12 13 14 15 16
logical_and	7 8 9 10 11 12 13 14 15 16
logical_or	7 8 9 10 11 12 13 14 15 16
logical_xor	7 8 9 10 11 12 13 14 15 16
logsumexp	7 8 9 10 11 12 13 14 15 16
lstm	7 8 9 10 11 12 13 14 15 16
lstm_cell	7 8 9 10 11 12 13 14 15 16
lt	7 8 9 10 11 12 13 14 15 16
masked_fill	7 8 9 10 11 12 13 14 15 16
masked_scatter	11 12 13 14 15 16
masked_select	11 12 13 14 15 16
matmul	7 8 9 10 11 12 13 14 15 16
max	7 7 8 8 9 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
max_pool1d	7 8 9 10 11 12 13 14 15 16
max_pool1d_with_indices	7 8 9 10 11 12 13 14 15 16
max_pool2d	7 8 9 10 11 12 13 14 15 16
max_pool2d_with_indices	7 8 9 10 11 12 13 14 15 16
max_pool3d	7 8 9 10 11 12 13 14 15 16
max_pool3d_with_indices	7 8 9 10 11 12 13 14 15 16
maximum	7 8 9 10 11 12 13 14 15 16
mean	7 8 9 10 11 12 13 14 15 16
meshgrid	7 8 9 10 11 12 13 14 15 16
min	7 7 8 8 9 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
minimum	7 8 9 10 11 12 13 14 15 16
mish	7 8 9 10 11 12 13 14 15 16
mm	7 8 9 10 11 12 13 14 15 16
mul	7 8 9 10 10 11 11 12 12 13 13 14 14 15 15 16 16
multinomial	7 8 9 10 11 12 13 14 15 16
mv	7 8 9 10 11 12 13 14 15 16
nan_to_num	10 11 12 13 14 15 16
narrow	7 8 9 10 11 12 13 14 15 16
ne	7 8 9 10 11 12 13 14 15 16
neg	7 8 9 10 11 12 13 14 15 16
new_empty	7 8 9 10 11 12 13 14 15 16
new_full	7 8 9 10 11 12 13 14 15 16
new_ones	7 8 9 10 11 12 13 14 15 16
new_zeros	7 8 9 10 11 12 13 14 15 16
nll_loss	12 13 14 15 16
nll_loss2d	12 13 14 15 16
nll_loss_nd	12 13 14 15 16
nonzero	7 8 9 10 11 12 13 14 15 16
nonzero_numpy	7 8 9 10 11 12 13 14 15 16
norm	7 8 9 10 11 12 13 14 15 16
normal	11 12 13 14 15 16
numel	7 8 9 10 11 12 13 14 15 16
one_hot	7 8 9 10 11 12 13 14 15 16
ones	7 8 9 10 11 12 13 14 15 16
ones_like	7 8 9 10 11 12 13 14 15 16
outer	12 13 14 15 16
pad	7 8 9 10 11 12 13 14 15 16
pairwise_distance	7 8 9 10 11 12 13 14 15 16
permute	7 8 9 10 11 12 13 14 15 16
pixel_shuffle	7 8 9 10 11 12 13 14 15 16
pixel_unshuffle	7 8 9 10 11 12 13 14 15 16
pop	11 12 13 14 15 16
pow	7 8 9 10 11 12 13 14 15 16
prelu	7 8 9 10 11 12 13 14 15 16
prod	7 8 9 10 11 12 13 14 15 16
quantize_per_tensor	10 11 12 13 14 15 16
rand	7 8 9 10 11 12 13 14 15 16
rand_like	7 8 9 10 11 12 13 14 15 16
randn	7 8 9 10 11 12 13 14 15 16
randn_like	7 8 9 10 11 12 13 14 15 16
reciprocal	7 8 9 10 11 12 13 14 15 16
reflection_pad1d	7 8 9 10 11 12 13 14 15 16
reflection_pad2d	7 8 9 10 11 12 13 14 15 16
reflection_pad3d	7 8 9 10 11 12 13 14 15 16
relu	7 8 9 10 11 12 13 14 15 16
relu6	7 8 9 10 11 12 13 14 15 16
remainder	7 8 9 10 11 12 13 14 15 16
repeat	7 8 9 10 11 12 13 14 15 16
repeat_interleave	7 8 9 10 11 12 13 14 15 16
replication_pad1d	7 8 9 10 11 12 13 14 15 16
replication_pad2d	7 8 9 10 11 12 13 14 15 16
replication_pad3d	7 8 9 10 11 12 13 14 15 16
reshape	7 8 9 10 11 12 13 14 15 16
reshape_as	7 8 9 10 11 12 13 14 15 16
rnn_relu	7 8 9 10 11 12 13 14 15 16
rnn_tanh	7 8 9 10 11 12 13 14 15 16
roll	7 8 9 10 11 12 13 14 15 16
round	11 12 13 14 15 16
rrelu	7 8 9 10 11 12 13 14 15 16
rsqrt	7 8 9 10 11 12 13 14 15 16
rsub	7 8 9 10 11 12 13 14 15 16
scalar_tensor	7 8 9 10 11 12 13 14 15 16
scatter	7 8 9 10 11 12 13 14 15 16
scatter_add	7 8 9 10 11 12 13 14 15 16
select	7 8 9 10 11 12 13 14 15 16
selu	7 8 9 10 11 12 13 14 15 16
sigmoid	7 8 9 10 11 12 13 14 15 16
sign	7 8 9 10 11 12 13 14 15 16
silu	7 8 9 10 11 12 13 14 15 16
sin	7 8 9 10 11 12 13 14 15 16
size	7 8 9 10 11 12 13 14 15 16
slice	7 8 9 10 11 12 13 14 15 16
softmax	7 8 9 10 11 12 13 14 15 16
softplus	7 8 9 10 11 12 13 14 15 16
softshrink	7 8 9 10 11 12 13 14 15 16
sort	7 8 9 10 11 12 13 14 15 16
split	7 8 9 10 11 12 13 14 15 16
split_with_sizes	7 8 9 10 11 12 13 14 15 16
sqrt	7 8 9 10 11 12 13 14 15 16
square	7 8 9 10 11 12 13 14 15 16
squeeze	7 8 9 10 11 12 13 14 15 16
stack	7 8 9 10 11 12 13 14 15 16
std	7 8 9 10 11 12 13 14 15 16
std_mean	7 8 9 10 11 12 13 14 15 16
sub	7 8 9 10 11 12 13 14 15 16
sum	7 8 9 10 11 12 13 14 15 16
t	7 8 9 10 11 12 13 14 15 16
take	7 8 9 10 11 12 13 14 15 16
tan	7 8 9 10 11 12 13 14 15 16
tanh	7 8 9 10 11 12 13 14 15 16
tensor	7 8 9 10 11 12 13 14 15 16
tensordot	12 13 14 15 16
threshold	7 8 9 10 11 12 13 14 15 16
to	7 8 9 10 11 12 13 14 15 16
topk	7 8 9 10 11 12 13 14 15 16
transpose	7 8 9 10 11 12 13 14 15 16
tril	14 15 16
triu	14 15 16
true_divide	7 8 9 10 11 12 13 14 15 16
type_as	7 8 9 10 11 12 13 14 15 16
unbind	7 8 9 10 11 12 13 14 15 16
unfold	7 8 9 10 11 12 13 14 15 16
unique_dim	11 12 13 14 15 16
unsafe_chunk	7 8 9 10 11 12 13 14 15 16
unsafe_split	7 8 9 10 11 12 13 14 15 16
unsafe_split_with_sizes	7 8 9 10 11 12 13 14 15 16
unsqueeze	7 8 9 10 11 12 13 14 15 16
upsample_bicubic2d	11 12 13 14 15 16
upsample_bilinear2d	7 8 9 10 11 12 13 14 15 16
upsample_linear1d	7 8 9 10 11 12 13 14 15 16
upsample_nearest1d	7 8 9 10 11 12 13 14 15 16
upsample_nearest2d	7 8 9 10 11 12 13 14 15 16
upsample_nearest3d	7 8 9 10 11 12 13 14 15 16
upsample_trilinear3d	7 8 9 10 11 12 13 14 15 16
var	7 8 9 10 11 12 13 14 15 16
var_mean	7 8 9 10 11 12 13 14 15 16
view	7 8 9 10 11 12 13 14 15 16
view_as	7 8 9 10 11 12 13 14 15 16
where	7 8 9 10 11 12 13 14 15 16
zeros	7 8 9 10 11 12 13 14 15 16
zeros_like	7 8 9 10 11 12 13 14 15 16