MCPcopy Create free account

hub / github.com/VAST-AI-Research/HoloPart / functions

Functions124 in github.com/VAST-AI-Research/HoloPart

⨍ Functions 124◇ Types & classes 19
Methodfuse_qkv_projections
Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att
holopart/pipelines/pipeline_utils.py:40
Methodfuse_qkv_projections
Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att
holopart/models/transformers/triposg_transformer.py:656
Methodfuse_qkv_projections
Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value) are fused. For cross-att
holopart/models/autoencoders/autoencoder_kl_triposg.py:276
Methodguidance_scale
(self)
holopart/pipelines/pipeline_holopart.py:110
Methodinterrupt
(self)
holopart/pipelines/pipeline_holopart.py:126
Methodkl
(self, other: "DiagonalGaussianDistribution" = None, dims=(1, 2, 3))
holopart/models/autoencoders/vae.py:38
Methodmode
(self)
holopart/models/autoencoders/vae.py:68
Methodnll
( self, sample: torch.Tensor, dims: Tuple[int, ...] = [1, 2, 3] )
holopart/models/autoencoders/vae.py:57
Methodnum_timesteps
(self)
holopart/pipelines/pipeline_holopart.py:118
Methodquery_fn
(q, kv)
holopart/models/autoencoders/autoencoder_kl_triposg.py:208
Methodset_begin_index
Sets the begin index for the scheduler. This function should be run from pipeline before the inference. Args: begin_inde
holopart/schedulers/scheduling_rectified_flow.py:144
Methodset_default_attn_processor
Disables custom attention processors and sets the default attention implementation.
holopart/models/transformers/triposg_transformer.py:763
Methodset_default_attn_processor
Disables custom attention processors and sets the default attention implementation.
holopart/models/autoencoders/autoencoder_kl_triposg.py:383
Methodset_flash_processor
(self, flash_processor)
holopart/models/transformers/triposg_transformer.py:290
Methodset_topk
(self, topk)
holopart/models/autoencoders/autoencoder_kl_triposg.py:148
Functionset_transformer_attn_processor
( transformer: TripoSGDiTModel, set_self_attn_proc_func: Callable = default_set_attn_proc_func, se
holopart/models/transformers/__init__.py:15
Methodset_use_cross_attention
(self, use_cross_attention, use_cross_attention_2)
holopart/models/transformers/triposg_transformer.py:617
Methodstep_index
The index counter for current timestep. It will increase 1 after each scheduler step.
holopart/schedulers/scheduling_rectified_flow.py:130
Methodunfuse_qkv_projections
Disable QKV projection fusion if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip> Args:
holopart/pipelines/pipeline_utils.py:66
Methodunfuse_qkv_projections
Disables the fused QKV projection if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip>
holopart/models/transformers/triposg_transformer.py:684
Methodunfuse_qkv_projections
Disables the fused QKV projection if enabled. <Tip warning={true}> This API is 🧪 experimental. </Tip>
holopart/models/autoencoders/autoencoder_kl_triposg.py:304
Methodupdate_cross_attention
(self, name, cross_attention_dim, cross_attention_norm_type = None)
holopart/models/transformers/triposg_transformer.py:613
Methodvalidate
(variable)
holopart/utils/typing.py:52
Functionzoom_block
(block, scale_factor, order=3)
holopart/inference_utils.py:119
← previous101–124 of 124, ranked by callers