MCPcopy Create free account

hub / github.com/Xuehao-Gao/GUESS / functions

Functions912 in github.com/Xuehao-Gao/GUESS

⨍ Functions 912◇ Types & classes 142
↓ 88 callersMethodto
(self, *args, **kwargs)
mld/transforms/base.py:50
↓ 66 callersMethoddetach
(self)
mld/transforms/base.py:60
↓ 40 callersMethoditems
(self)
mld/transforms/base.py:46
↓ 24 callersMethodkeys
(self)
mld/transforms/base.py:38
↓ 24 callersFunctionl2_norm
(x1, x2, dim)
mld/models/metrics/utils.py:8
↓ 24 callersFunctionvariance
(x, T, dim)
mld/models/metrics/utils.py:12
↓ 23 callersMethodencode
( self, features: Tensor, lengths: Optional[List[int]] = None )
mld/models/architectures/mld_vae.py:127
↓ 23 callersMethodload_state_dict
(self, state_dict, strict=True)
mld/models/modeltype/base.py:117
↓ 22 callersMethodsave
(self, out_path)
mld/render/video.py:64
↓ 21 callersMethodupdate
( self, mm_motion_embeddings: Tensor, lengths: List[int], )
mld/models/metrics/mm.py:53
↓ 19 callersMethodto
(self, device)
mld/models/operator/position_encoding.py:21
↓ 17 callersMethodfeats2joints
(self, features)
mld/data/Kit.py:43
↓ 17 callersFunctioninstantiate_from_config
(config)
mld/config.py:24
↓ 15 callersMethod_update_loss
(self, loss: str, outputs, inputs)
mld/models/losses/mld.py:164
↓ 12 callersMethod_diffusion_reverse
(self, encoder_hidden_states, lengths=None, scale='s1', coarse_motion=None, reverse_in_test_flage=False)
mld/models/modeltype/mld.py:323
↓ 12 callersMethod_update_loss
(self, loss: str, outputs, inputs)
mld/models/losses/tmost.py:172
↓ 11 callersMethoddecode
(self, z: Tensor, lengths: List[int], scale: str)
mld/models/architectures/mld_vae.py:197
↓ 10 callersMethodnormalize
(self, features: Tensor)
mld/transforms/rots2joints/base.py:26
↓ 10 callersFunctionqrot_np
(q, v)
mld/data/humanml/common/quaternion.py:134
↓ 9 callersFunction_sqn
(arr)
mld/models/metrics/utils.py:597
↓ 9 callersMethod_update_loss
(self, loss: str, outputs, inputs)
mld/models/losses/temos.py:156
↓ 8 callersFunction_conv_filter
convert patch embedding weight from manual patchify + linear proj to conv
mld/models/architectures/vision_transformer.py:561
↓ 8 callersFunction_get_clones
(module, N)
mld/models/operator/cross_attention.py:387
↓ 8 callersFunctionqmul_np
(q, r)
mld/data/humanml/common/quaternion.py:128
↓ 8 callersFunctionremove_padding
(tensors, lengths)
mld/models/tools/tools.py:3
↓ 8 callersMethodwrite
(self, buf)
mld/tools/logging.py:22
↓ 6 callersMethod__init__
(self, d_model=512, nhead=8, num_encoder_layers=6, num_decoder_layers=6, dim_feedforward=2048
mld/models/operator/cross_attention.py:129
↓ 6 callersFunctioncalculate_activation_statistics
Params: -- activation: num_samples x dim_feat Returns: -- mu: dim_feat -- sigma: dim_feat x dim_feat
mld/models/metrics/utils.py:78
↓ 6 callersFunctioncollate_tensors
(batch)
mld/data/utils.py:12
↓ 6 callersMethodtransform
(self, joints: Tensor, lengths)
mld/models/metrics/compute.py:133
↓ 6 callersMethodwith_pos_embed
(self, tensor, pos: Optional[Tensor])
mld/models/operator/cross_attention.py:320
↓ 5 callersMethod__init__
(self)
mld/transforms/multi_scale_init.py:86
↓ 5 callersMethod__init__
(self, in_features, hidden_features=None, out_features=None
mld/models/architectures/vision_transformer.py:145
↓ 5 callersFunctionas_minutes
(s)
mld/data/humanml/utils/utils.py:39
↓ 5 callersMethodinverse_kinematics_np
(self, joints, face_joint_idx, smooth_forward=False)
mld/data/humanml/common/skeleton.py:55
↓ 5 callersFunctionlengths_to_mask
(lengths: List[int], device: torch.device, max_len: int = None)
mld/utils/temos_utils.py:10
↓ 5 callersFunctionqinv_np
(q)
mld/data/humanml/common/quaternion.py:23
↓ 5 callersFunctionqmul
Multiply quaternion(s) q with quaternion(s) r. Expects two equally-sized tensors of shape (*, 4), where * denotes any number of dimensions.
mld/data/humanml/common/quaternion.py:33
↓ 5 callersFunctionslice_or_none
(data, cslice)
mld/transforms/rots2joints/smplh.py:10
↓ 4 callersMethod__init__
Positional Encoding with maximum length max_len :param size: :param max_len: :param dropout:
mld/models/architectures/tools/transformer_layers.py:136
↓ 4 callersMethod__init__
(self, num_channels: int, flip_sin_to_cos: bool, downscale_freq_shift: float)
mld/models/architectures/tools/embeddings.py:309
↓ 4 callersMethod__init__
(self, tensors, mask: Optional[Tensor])
mld/models/operator/position_encoding.py:17
↓ 4 callersFunction_copysign
Return a tensor where each element has the absolute value taken from the, corresponding element of a, with sign taken from the corresponding
mld/tools/geometry.py:83
↓ 4 callersFunction_copysign
Return a tensor where each element has the absolute value taken from the, corresponding element of a, with sign taken from the corresponding
mld/utils/rotation_conversions.py:68
↓ 4 callersFunction_copysign
Return a tensor where each element has the absolute value taken from the, corresponding element of a, with sign taken from the corresponding
mld/data/a2m/utils/rotation_conversions.py:68
↓ 4 callersMethod_diffusion_process
heavily from https://github.com/huggingface/diffusers/blob/main/examples/dreambooth/train_dreambooth.py
mld/models/modeltype/mld.py:450
↓ 4 callersFunction_sqrt_positive_part
Returns torch.sqrt(torch.max(0, x)) but with a zero subgradient where x is 0.
mld/tools/geometry.py:101
↓ 4 callersFunction_sqrt_positive_part
Returns torch.sqrt(torch.max(0, x)) but with a zero subgradient where x is 0.
mld/utils/rotation_conversions.py:86
↓ 4 callersFunction_sqrt_positive_part
Returns torch.sqrt(torch.max(0, x)) but with a zero subgradient where x is 0.
mld/data/a2m/utils/rotation_conversions.py:86
↓ 4 callersMethodbackward
(ctx, grad_output)
mld/models/operator/conv2d_gradfix.py:154
↓ 4 callersFunctionbuild_position_encoding
(N_steps, position_embedding="sine", embedding_dim="1D
mld/models/operator/position_encoding.py:163
↓ 4 callersFunctioncalculate_activation_statistics_np
Params: -- activation: num_samples x dim_feat Returns: -- mu: dim_feat -- sigma: dim_feat x dim_feat
mld/models/metrics/utils.py:92
↓ 4 callersFunctioncalculate_diversity_multimodality
(activations, labels, num_labels,
mld/models/metrics/utils.py:410
↓ 4 callersFunctioncalculate_diversity_np
(activation, diversity_times)
mld/models/metrics/utils.py:230
↓ 4 callersFunctioncalculate_fid
(statistics_1, statistics_2)
mld/models/metrics/utils.py:455
↓ 4 callersFunctionconv2d_gradfix
( transpose, weight_shape, stride, padding, output_padding, dilation, groups )
mld/models/operator/conv2d_gradfix.py:104
↓ 4 callersFunctionensure_tuple
(xs, ndim)
mld/models/operator/conv2d_gradfix.py:95
↓ 4 callersMethodget
(self, key, default=None)
mld/transforms/base.py:32
↓ 4 callersMethodget_offsets_joints_batch
(self, joints)
mld/data/humanml/common/skeleton.py:33
↓ 4 callersFunctionparse_args
(phase="train")
mld/config.py:34
↓ 4 callersFunctionpool_job_with_desc
(pool, iterator, desc, max_, save_path_format, isij)
mld/render/visualize.py:263
↓ 4 callersFunctionqinv
(q)
mld/data/humanml/common/quaternion.py:16
↓ 4 callersFunctionqnormalize
(q)
mld/data/humanml/common/quaternion.py:28
↓ 4 callersFunctionqrot
Rotate vector(s) v about the rotation described by quaternion(s) q. Expects a tensor of shape (*, 4) for q and a tensor of shape (*, 3) for v
mld/data/humanml/common/quaternion.py:54
↓ 4 callersFunctionrecover_from_ric
(data, joints_num)
mld/data/humanml/scripts/motion_process.py:415
↓ 4 callersFunctionremove_padding
(tensors, lengths)
mld/utils/temos_utils.py:24
↓ 4 callersMethodreshape_T
(self, x, seqlen=1, inverse=False)
mld/models/architectures/vision_transformer.py:259
↓ 4 callersFunctionsphere
(r, t, mat)
mld/render/blender/joints.py:234
↓ 4 callersFunctiontrunc_normal_
r"""Fills the input Tensor with values drawn from a truncated normal distribution. The values are effectively drawn from the normal distributi
mld/models/architectures/vision_transformer.py:122
↓ 4 callersMethodupdate
(self, newroot)
mld/render/blender/camera.py:46
↓ 3 callersFunctionG
(z)
mld/models/tools/hessian_penalty.py:126
↓ 3 callersFunctionT
(x)
mld/transforms/joints2jfeats/tools.py:10
↓ 3 callersMethod__init__
(self, cfg, **kwargs)
mld/models/architectures/vposert_vae.py:29
↓ 3 callersMethod_get_item_data_index
(self, data_index)
mld/data/a2m/dataset.py:149
↓ 3 callersMethod_get_pos_ohot
(self, pos)
mld/data/humanml/utils/word_vectorizer.py:53
↓ 3 callersMethodallsplit_epoch_end
(self, split: str, outputs)
mld/models/modeltype/base.py:42
↓ 3 callersMethodallsplit_step
(self, split: str, batch, batch_idx)
mld/models/modeltype/mld.py:1002
↓ 3 callersFunctionbody_fitting_loss_3d
Loss function for body fitting
mld/transforms/joints2rots/customloss.py:128
↓ 3 callersFunctionbody_material
(r, g, b, a=1, name="body", oldrender=True)
mld/render/blender/materials.py:60
↓ 3 callersFunctioncalculate_frechet_distance_np
Numpy implementation of the Frechet Distance. The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1) and X_2 ~ N(mu_2, C_2
mld/models/metrics/utils.py:161
↓ 3 callersFunctionclear_material
(material)
mld/render/blender/materials.py:4
↓ 3 callersFunctioncont6d_to_matrix
(cont6d)
mld/data/humanml/common/quaternion.py:320
↓ 3 callersFunctioncreate_logger
(cfg, phase='train')
mld/utils/logger.py:9
↓ 3 callersFunctiondelete_objs
(names)
mld/render/blender/tools.py:37
↓ 3 callersMethoddevice
(self)
mld/transforms/base.py:57
↓ 3 callersFunctionfloor_mat
(name="floor_mat", color=(0.1, 0.1, 0.1, 1), roughness=0.127451)
mld/render/blender/materials.py:100
↓ 3 callersMethodforward_spatial
(self, q, k, v)
mld/models/architectures/vision_transformer.py:286
↓ 3 callersMethodforward_temporal
(self, q, k, v, seqlen=8)
mld/models/architectures/vision_transformer.py:296
↓ 3 callersFunctionget_datasets
(cfg, logger=None, phase="train")
mld/data/get_data.py:77
↓ 3 callersFunctionget_model
(cfg, datamodule, phase="train")
mld/models/get_model.py:4
↓ 3 callersMethodget_root
(self, index)
mld/render/blender/joints.py:71
↓ 3 callersMethodlabel_to_action
(self, label)
mld/data/a2m/dataset.py:61
↓ 3 callersFunctionlistify
If x is already a list, do nothing. Otherwise, wrap x in a list.
mld/models/tools/hessian_penalty.py:108
↓ 3 callersFunctionload_and_clean_data
(path)
prepare/merge_smplh_mano.py:56
↓ 3 callersFunctionload_anim
(path, timesize=None)
mld/render/visualize.py:26
↓ 3 callersMethodmask_cond
(self, output, force=False)
mld/models/architectures/mld_denoiser.py:260
↓ 3 callersFunctionmatrix_of_angles
(cos, sin, inv=False, dim=2)
mld/tools/geometry.py:38
↓ 3 callersFunctionmatrix_to
(newtype, rotations)
mld/utils/temos_utils.py:65
↓ 3 callersMethodnormalize
(self, features: Tensor)
mld/transforms/joints2jfeats/base.py:26
↓ 3 callersFunctionqbetween_np
find the quaternion used to rotate v0 to v1
mld/data/humanml/common/quaternion.py:400
next →1–100 of 912, ranked by callers