Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/Yasoz/DiffTraj
/ functions
Functions
61 in github.com/Yasoz/DiffTraj
⨍
Functions
61
◇
Types & classes
12
↓ 16 callers
Method
format
(self, record)
utils/logger.py:61
↓ 7 callers
Method
__init__
(self, config)
utils/Traj_UNet.py:239
↓ 5 callers
Function
nonlinearity
(x)
utils/Traj_UNet.py:75
↓ 4 callers
Function
Normalize
(in_channels)
utils/Traj_UNet.py:80
↓ 3 callers
Method
state_dict
(self)
utils/EMA.py:48
↓ 2 callers
Function
compute_alpha
(beta, t)
utils/utils.py:60
↓ 2 callers
Function
gather
Gather consts for $t$ and reshape to feature map shape
main.py:27
↓ 2 callers
Function
gather
Gather consts for $t$ and reshape to feature map shape :param consts: (N, 1, 1) :param t: (N, H, W) :return: (N, H, W)
utils/utils.py:39
↓ 2 callers
Method
load_state_dict
(self, state_dict)
utils/EMA.py:51
↓ 1 callers
Function
Normalize
(in_channels)
utils/module.py:32
↓ 1 callers
Method
__init__
(self, name="default", colorize=False, log_path=None,
utils/logger.py:69
↓ 1 callers
Method
_process
(msg, loglevel, colorize)
utils/logger.py:36
↓ 1 callers
Method
add_handler
(self, stream)
utils/logger.py:130
↓ 1 callers
Method
clear_handlers
(self)
utils/logger.py:144
↓ 1 callers
Method
ema
(self, module)
utils/EMA.py:27
↓ 1 callers
Function
get_timestep_embedding
(timesteps, embedding_dim)
utils/Traj_UNet.py:9
↓ 1 callers
Function
log_info
output the information about model
utils/logger.py:166
↓ 1 callers
Function
main
(config, logger, exp_dir)
main.py:33
↓ 1 callers
Function
q_xt_x0
(x0, t)
main.py:36
↓ 1 callers
Method
register
(self, module)
utils/EMA.py:11
↓ 1 callers
Method
setLevel
(self, level: Union[str, int])
utils/logger.py:122
↓ 1 callers
Method
update
(self, module)
utils/EMA.py:18
Method
__init__
(self, embedding_dim)
utils/Traj_UNet.py:24
Method
__init__
(self, embedding_dim=128, hidden_dim=256)
utils/Traj_UNet.py:37
Method
__init__
(self, in_channels, with_conv=True)
utils/Traj_UNet.py:88
Method
__init__
(self, in_channels, with_conv=True)
utils/Traj_UNet.py:108
Method
__init__
(self, in_channels, out_channels=None, conv_shortcut=False,
utils/Traj_UNet.py:130
Method
__init__
(self, in_channels)
utils/Traj_UNet.py:191
Method
__init__
(self, config)
utils/Traj_UNet.py:403
Method
__init__
(self, colorize=False, *args, **kwargs)
utils/logger.py:31
Method
__init__
(self, in_channels)
utils/module.py:40
Method
__init__
(self, mu=0.999)
utils/EMA.py:7
Function
distance
Calculate the great circle distance between two points on the earth (specified in decimal degrees)
utils/utils.py:90
Function
divide_grids
(boundary, grids_num)
utils/utils.py:77
Method
ema_copy
(self, module)
utils/EMA.py:34
Function
extract
(input, t, x)
utils/helper.py:8
Method
forward
(self, x)
utils/Traj_UNet.py:28
Method
forward
(self, attr)
utils/Traj_UNet.py:50
Method
forward
(self, x)
utils/Traj_UNet.py:98
Method
forward
(self, x)
utils/Traj_UNet.py:119
Method
forward
(self, x, temb)
utils/Traj_UNet.py:170
Method
forward
(self, x)
utils/Traj_UNet.py:217
Method
forward
(self, x, t, extra_embed=None)
utils/Traj_UNet.py:348
Method
forward
(self, x, t, attr)
utils/Traj_UNet.py:415
Method
forward
(self, x)
utils/module.py:66
Method
func_wrapper
(*args, **kwargs)
utils/logger.py:110
Method
get_handlers
(self)
utils/logger.py:148
Function
get_timestep_embedding
This matches the implementation in Denoising Diffusion Probabilistic Models: From Fairseq. Build sinusoidal embeddings. This matches
utils/module.py:6
Method
inner_formatter
(self)
utils/logger.py:162
Method
inner_logger
(self)
utils/logger.py:154
Method
inner_stream_handler
(self)
utils/logger.py:158
Method
log_function
(self)
utils/logger.py:108
Function
make_beta_schedule
(schedule='linear', n_timesteps=1000, start=1e-5, end=1e-2)
utils/helper.py:14
Function
nonlinearity
(x)
utils/module.py:27
Function
p_xt
(xt, noise, t, next_t, beta, eta=0)
utils/utils.py:66
Function
q_xt_x0
(x0, t, alpha_bar)
utils/utils.py:50
Method
remove_handler
(self, stream)
utils/logger.py:137
Function
resample_trajectory
Resamples a trajectory to a new length. Parameters: x (np.ndarray): original trajectory, shape (N, 2) length (int): length o
utils/utils.py:6
Function
sample_batch
(size, noise=1.0)
utils/helper.py:4
Function
time_warping
Resamples a trajectory to a new length.
utils/utils.py:26
Method
wrapper
(func)
utils/logger.py:109