Types & classes — github.com/Sin3DM/Sin3DM

CodeHub MCPcopy Create free account

hub / github.com/Sin3DM/Sin3DM / types & classes

Types & classes50 in github.com/Sin3DM/Sin3DM

⨍ Functions 339◇ Types & classes 50

↓ 8 callersClassTriplaneConv

src/diffusion/unet_triplane.py:21

↓ 7 callersClassTriplaneGroupResnetBlock

src/encoding/blocks.py:189

↓ 6 callersClassDecoderMLPSkipConcat

src/encoding/blocks.py:65

↓ 6 callersClassSiLU

src/encoding/blocks.py:94

↓ 6 callersClassTimestepEmbedSequential

A sequential module that passes timestep embeddings to the children that support it as an extra input.

src/diffusion/unet_triplane.py:160

↓ 4 callersClassSiLU

src/diffusion/nn.py:12

↓ 4 callersClassTriplaneDownsample2x

src/diffusion/unet_triplane.py:127

↓ 4 callersClassTriplaneNorm

src/diffusion/unet_triplane.py:63

↓ 4 callersClassTriplaneResBlock

A residual block that can optionally change the number of channels. :param channels: the number of input channels. :param emb_channels:

src/diffusion/unet_triplane.py:175

↓ 4 callersClassTriplaneSiLU

src/diffusion/unet_triplane.py:87

↓ 4 callersClassTriplaneUpsample2x

src/diffusion/unet_triplane.py:106

↓ 2 callersClassDecoderMLP

src/encoding/blocks.py:45

↓ 2 callersClassHumanOutputFormat

src/diffusion/logger.py:36

↓ 2 callersClassInceptionV3

Pretrained InceptionV3 network returning feature maps

evaluation/inception.py:6

↓ 2 callersClassLPIPS

evaluation/lpips.py:52

↓ 2 callersClassShapeAutoEncoder

src/encoding/model.py:15

↓ 2 callersClassSinusoidalEncoder

Sinusoidal Positional Encoder used in Nerf.

src/encoding/blocks.py:7

↓ 2 callersClassUniformSampler

src/diffusion/resample.py:61

↓ 1 callersClassAlexNet

evaluation/lpips.py:23

↓ 1 callersClassAutoEncoderGroupPBR

src/encoding/networks.py:227

↓ 1 callersClassAutoEncoderGroupSkip

src/encoding/networks.py:124

↓ 1 callersClassAutoEncoderGroupV3

src/encoding/networks.py:21

↓ 1 callersClassCSVOutputFormat

src/diffusion/logger.py:113

↓ 1 callersClassConv1x1

evaluation/lpips.py:41

↓ 1 callersClassEmptyArgs

↓ 1 callersClassGaussianDiffusion

Utilities for training and sampling diffusion models. Ported directly from here, and then adapted over time to further experimentation.

src/diffusion/gaussian_diffusion.py:102

↓ 1 callersClassGroupNorm32

src/diffusion/nn.py:17

↓ 1 callersClassJSONOutputFormat

src/diffusion/logger.py:98

↓ 1 callersClassLogger

src/diffusion/logger.py:332

↓ 1 callersClassLossSecondMomentResampler

src/diffusion/resample.py:124

↓ 1 callersClassMeshSampler

data/mesh_sampler.py:11

↓ 1 callersClassMeshSampler

data/mesh_sampler_pbr.py:13

↓ 1 callersClassMixedPrecisionTrainer

src/diffusion/fp16_util.py:148

↓ 1 callersClassSpacedDiffusion

A diffusion process which can skip steps in a base diffusion process. :param use_timesteps: a collection (sequence or set) of timesteps from

src/diffusion/respace.py:63

↓ 1 callersClassTensorBoardOutputFormat

Dumps key/value pairs into TensorBoard's numeric format.

src/diffusion/logger.py:150

↓ 1 callersClassTrainLoop

src/diffusion/train_util.py:26

↓ 1 callersClassTriplaneUNetModelSmall

The full UNet model with attention and timestep embedding. :param in_channels: channels in the input Tensor. :param model_channels: base

src/diffusion/unet_triplane.py:315

↓ 1 callersClassTriplaneUNetModelSmallRaw

The full UNet model with attention and timestep embedding. :param in_channels: channels in the input Tensor. :param model_channels: base

src/diffusion/unet_triplane.py:513

↓ 1 callersClass_WrappedModel

src/diffusion/respace.py:116

↓ 1 callersClassclassifier

evaluation/classifier3D.py:6

ClassCheckpointFunction

src/diffusion/nn.py:142

src/diffusion/logger.py:26

ClassLossAwareSampler

src/diffusion/resample.py:70

src/diffusion/gaussian_diffusion.py:90

ClassModelMeanType

Which type of output the model predicts.

src/diffusion/gaussian_diffusion.py:66

ClassModelVarType

What is used as the model's output variance. The LEARNED_RANGE option has been added to allow the model to predict values between FIXED_

src/diffusion/gaussian_diffusion.py:76

ClassResnetBlock

src/encoding/blocks.py:108

ClassScheduleSampler

A distribution over timesteps in the diffusion process, intended to reduce variance of the objective. By default, samplers perform unbia

src/diffusion/resample.py:23

src/diffusion/logger.py:31

ClassTimestepBlock

Any module where forward() takes timestep embeddings as a second argument.

src/diffusion/unet_triplane.py:148