ding.torch_utils¶
loss¶
Please refer to ding/torch_utils/loss for more details.
ContrastiveLoss¶
LabelSmoothCELoss¶
SoftFocalLoss¶
build_ce_criterion¶
MultiLogitsLoss¶
network.activation¶
Please refer to ding/torch_utils/network/activation for more details.
Lambda¶
GLU¶
Swish¶
GELU¶
build_activation¶
network.diffusion¶
Please refer to ding/torch_utils/network/diffusion for more details.
extract¶
cosine_beta_schedule¶
apply_conditioning¶
DiffusionConv1d¶
SinusoidalPosEmb¶
Residual¶
LayerNorm¶
PreNorm¶
LinearAttention¶
ResidualTemporalBlock¶
DiffusionUNet1d¶
TemporalValue¶
network.dreamer¶
Please refer to ding/torch_utils/network/dreamer for more details.
Conv2dSame¶
DreamerLayerNorm¶
DenseHead¶
ActionHead¶
SampleDist¶
OneHotDist¶
TwoHotDistSymlog¶
SymlogDist¶
ContDist¶
Bernoulli¶
network.gtrxl¶
Please refer to ding/torch_utils/network/gtrxl for more details.
PositionalEmbedding¶
GRUGatingUnit¶
Memory¶
AttentionXL¶
GatedTransformerXLLayer¶
GTrXL¶
network.gumbel_softmax¶
Please refer to ding/torch_utils/network/gumbel_softmax for more details.
GumbelSoftmax¶
network.merge¶
Please refer to ding/torch_utils/network/merge for more details.
BilinearGeneral¶
TorchBilinearCustomized¶
FiLM¶
GatingType¶
SumMerge¶
VectorMerge¶
network.nn_module¶
Please refer to ding/torch_utils/network/nn_module for more details.
weight_init¶
sequential_pack¶
conv1d_block¶
conv2d_block¶
deconv2d_block¶
fc_block¶
normed_linear¶
normed_conv2d¶
MLP¶
ChannelShuffle¶
one_hot¶
NearestUpsample¶
BilinearUpsample¶
binary_encode¶
NoiseLinearLayer¶
noise_block¶
NaiveFlatten¶
network.normalization¶
Please refer to ding/torch_utils/network/normalization for more details.
build_normalization¶
network.popart¶
Please refer to ding/torch_utils/network/popart for more details.
PopArt¶
network.res_block¶
Please refer to ding/torch_utils/network/res_block for more details.
ResBlock¶
ResFCBlock¶
network.resnet¶
Please refer to ding/torch_utils/network/resnet for more details.
to_2tuple¶
get_same_padding¶
pad_same¶
avg_pool2d_same¶
AvgPool2dSame¶
create_classifier¶
ClassifierHead¶
create_attn¶
get_padding¶
BasicBlock¶
Bottleneck¶
downsample_conv¶
downsample_avg¶
drop_blocks¶
make_blocks¶
ResNet¶
resnet18¶
network.rnn¶
Please refer to ding/torch_utils/network/rnn for more details.
is_sequence¶
sequence_mask¶
LSTMForwardWrapper¶
LSTM¶
PytorchLSTM¶
GRU¶
get_lstm¶
network.scatter_connection¶
Please refer to ding/torch_utils/network/scatter_connection for more details.
shape_fn_scatter_connection¶
ScatterConnection¶
network.soft_argmax¶
Please refer to ding/torch_utils/network/soft_argmax for more details.
SoftArgmax¶
network.transformer¶
Please refer to ding/torch_utils/network/transformer for more details.
Attention¶
TransformerLayer¶
Transformer¶
ScaledDotProductAttention¶
backend_helper¶
Please refer to ding/torch_utils/backend_helper for more details.
enable_tf32¶
checkpoint_helper¶
Please refer to ding/torch_utils/checkpoint_helper for more details.
build_checkpoint_helper¶
CheckpointHelper¶
CountVar¶
auto_checkpoint¶
data_helper¶
Please refer to ding/torch_utils/data_helper for more details.
to_device¶
to_dtype¶
to_tensor¶
to_ndarray¶
to_list¶
tensor_to_list¶
to_item¶
same_shape¶
LogDict¶
build_log_buffer¶
CudaFetcher¶
get_tensor_data¶
unsqueeze¶
squeeze¶
get_null_data¶
zeros_like¶
dataparallel¶
Please refer to ding/torch_utils/dataparallel for more details.
DataParallel¶
distribution¶
Please refer to ding/torch_utils/distribution for more details.
Pd¶
CategoricalPd¶
CategoricalPdPytorch¶
lr_scheduler¶
Please refer to ding/torch_utils/lr_scheduler for more details.
get_lr_ratio¶
cos_lr_scheduler¶
math_helper¶
Please refer to ding/torch_utils/math_helper for more details.
cov¶
metric¶
Please refer to ding/torch_utils/metric for more details.
levenshtein_distance¶
hamming_distance¶
model_helper¶
Please refer to ding/torch_utils/model_helper for more details.
get_num_params¶
nn_test_helper¶
Please refer to ding/torch_utils/nn_test_helper for more details.
is_differentiable¶
optimizer_helper¶
Please refer to ding/torch_utils/optimizer_helper for more details.
calculate_grad_norm¶
calculate_grad_norm_without_bias_two_norm¶
grad_ignore_norm¶
grad_ignore_value¶
Adam¶
RMSprop¶
PCGrad¶
configure_weight_decay¶
parameter¶
Please refer to ding/torch_utils/parameter for more details.
NonegativeParameter¶
TanhParameter¶
reshape_helper¶
Please refer to ding/torch_utils/reshape_helper for more details.