Functions — github.com/baaivision/Uni3D

CodeHub MCPcopy Create free account

hub / github.com/baaivision/Uni3D / functions

Functions189 in github.com/baaivision/Uni3D

⨍ Functions 189◇ Types & classes 21

↓ 1 callersFunctionwhitespace_clean

utils/tokenizer.py:58

utils/utils.py:242

(self, texts, context_length=77)

utils/tokenizer.py:136

data/utils/utils.py:242

Method__contains__

data/utils/registry.py:59

Method__getitem__

data/datasets.py:189

Method__getitem__

data/datasets.py:239

Method__getitem__

data/datasets.py:353

Method__getitem__

data/datasets.py:539

(self, name, fmt=':f')

(self, num_batches, meters, prefix="")

utils/optim.py:46

(self, sigma=[.1, 2.])

utils/utils.py:239

(self, bpe_path: str = default_bpe())

utils/tokenizer.py:65

data/datasets.py:160

data/datasets.py:213

data/datasets.py:268

data/datasets.py:470

(self, args, tokenizer, dataset_type, train_transform=None)

data/datasets.py:676

(self, sigma=[.1, 2.])

data/utils/utils.py:239

(self, name, build_func=None, parent=None, scope=None)

data/utils/registry.py:32

(self, prob, exclude_first_token=True)

models/point_encoder.py:57

(self, encoder_channel)

models/point_encoder.py:130

(self, point_transformer, args)

models/point_encoder.py:162

(self, point_encoder)

models/uni3d.py:10

models/losses.py:8

data/datasets.py:186

data/datasets.py:228

data/datasets.py:465

data/datasets.py:583

data/utils/registry.py:56

data/utils/registry.py:62

Function_lr_adjuster

utils/scheduler.py:14

Method_register

data/utils/registry.py:238

Functionall_gather_batch

Performs all_gather operation on the provided tensors.

utils/utils.py:155

Functionall_gather_batch

Performs all_gather operation on the provided tensors.

data/utils/utils.py:155

Functionall_gather_batch_with_grad

Performs all_gather operation on the provided tensors. Graph remains connected for backward grad computation.

utils/utils.py:200

Functionall_gather_batch_with_grad

Performs all_gather operation on the provided tensors. Graph remains connected for backward grad computation.

data/utils/utils.py:200

Functionaugment_pc

data/utils/data.py:79

data/utils/utils.py:194

Functionbuild_from_cfg

Build a module from config dict. Args: cfg (edict): Config dict. It should at least contain the key "NAME". registry (:obj:`Regist

data/utils/registry.py:246

Functioncfg_from_yaml_file

utils/utils.py:32

Functioncfg_from_yaml_file

data/utils/utils.py:32

data/utils/registry.py:120

Functioncosine_lr

(optimizer, base_lr, warmup_length, steps)

utils/scheduler.py:13

Functioncosine_scheduler

(base_value, final_value, epochs, niter_per_ep, warmup_epochs=0, start_warmup_value=0)

utils/utils.py:222

Functioncosine_scheduler

(base_value, final_value, epochs, niter_per_ep, warmup_epochs=0, start_warmup_value=0)

data/utils/utils.py:222

Functioncreate_uni3d

models/uni3d.py:36

Functioncustomized_collate_fn

r"""Puts each data field into a tensor with outer dimension batch size

data/datasets.py:597

Functionfarthest_point_sample

Input: xyz: pointcloud data, [N, D] npoint: number of samples Return: centroids: sampled pointcloud index, [npoint, D

data/datasets.py:36

utils/utils.py:188

data/utils/utils.py:188

models/point_encoder.py:64

input: B N 3 --------------------------- output: B G M 3 center : B G 3

models/point_encoder.py:99

point_groups : B G N 3 ----------------- feature_global : B G C

models/point_encoder.py:145

(self, pts, colors)

models/point_encoder.py:193

(self, pc, text, image)

models/uni3d.py:21

(self, outputs, masks)

models/losses.py:13

Functionget_config

(args, logger=None)

data/utils/config.py:47

Functionget_dataset

(train_transform, tokenizer, args, dataset_name=None)

data/utils/utils.py:247

Functionget_filter_loss

models/uni3d.py:30

Methodget_layer_id

(self, var_name)

utils/optim.py:52

Functionget_metric_names

models/uni3d.py:33

Functionget_model

utils/utils.py:42

Functionget_model

data/utils/utils.py:42

Functionget_root_logger

Get root logger and add a keyword filter to it. The logger will be initialized if it has not been initialized. By default a StreamHandler will

data/utils/logger.py:6

Functionget_world_size

utils/distributed.py:20

Functioninit_distributed_mode

utils/utils.py:105

Functioninit_distributed_mode

data/utils/utils.py:105

Functionjitter_point_cloud

Randomly jitter points. jittering is per point. Input: BxNx3 array, original batch of point clouds Return: BxNx3

data/datasets.py:114

Functionlog_args_to_file

(args, pre='args', logger=None)

data/utils/config.py:6

Functionlog_config_to_file

(cfg, pre='cfg', logger=None)

data/utils/config.py:10

Methodmodule_dict

data/utils/registry.py:116

data/utils/registry.py:108

Functionpil_loader

data/datasets.py:23

Functionprint_rank_0

If distributed is initialized, print only on rank 0.

utils/distributed.py:5

Functionrandom_rotate_z

data/utils/data.py:3

Methodrandom_sample

(self, pc, num)

data/datasets.py:343

Methodrandom_sample

(self, pc, num)

data/datasets.py:529

Methodregister_module

Register a module. A record will be added to `self._module_dict`, whose key is the class name or the specified name, and value is the

data/utils/registry.py:193

Functionsave_on_master

(state, is_best, output_dir)

utils/utils.py:89

Functionsave_on_master

(state, is_best, output_dir)

data/utils/utils.py:89

Functionsave_on_master_tmp

(state, is_best, output_dir)

utils/utils.py:97

Functionsave_on_master_tmp

(state, is_best, output_dir)

data/utils/utils.py:97

Functionscaled_all_reduce

Performs the scaled all_reduce operation on the provided tensors. The input tensors are modified in-place. Currently supports only the sum red

utils/utils.py:130

Functionscaled_all_reduce

Performs the scaled all_reduce operation on the provided tensors. The input tensors are modified in-place. Currently supports only the sum red

data/utils/utils.py:130

data/utils/registry.py:112

Functionwarmup_step_lr

(optimizer, args, decay_t=500, decay_rate=0.8)

utils/scheduler.py:48

← previous101–189 of 189, ranked by callers