Functions — github.com/OliverRensu/ARM

hub / github.com/OliverRensu/ARM / functions

Functions161 in github.com/OliverRensu/ARM

⨍ Functions 161◇ Types & classes 24

↓ 36 callersFunctionprint

(*args, **kwargs)

Finetuning/util/misc.py:176

↓ 27 callersFunctionprint

(*args, **kwargs)

utils/misc.py:171

↓ 9 callersMethodload_state_dict

(self, state_dict)

Finetuning/util/misc.py:276

↓ 8 callersMethodupdate

(self, **kwargs)

Finetuning/util/misc.py:91

↓ 7 callersMethodload_state_dict

(self, state_dict)

utils/misc.py:261

↓ 6 callersMethod__init__

(self, in_features, hidden_features=None, out_features=None, act_layer=nn.GELU, drop=0.)

models_pretrain.py:36

↓ 6 callersMethodstate_dict

(self)

utils/misc.py:258

↓ 5 callersFunctionevaluate

(data_loader, model, device)

Finetuning/engine_finetune.py:99

↓ 4 callersMethodupdate

(self, **kwargs)

utils/misc.py:86

↓ 3 callersMethod__init__

(self, img_size=224, patch_size=16, stride=16,

Finetuning/models_mamba.py:204

↓ 3 callersFunctionis_dist_avail_and_initialized

()

utils/misc.py:182

↓ 3 callersFunctionis_dist_avail_and_initialized

()

Finetuning/util/misc.py:187

↓ 3 callersMethodstate_dict

(self)

Finetuning/util/misc.py:273

↓ 2 callersFunctionbuild_dataset

(is_train, args)

Finetuning/util/datasets.py:20

↓ 2 callersFunctionget_1d_sincos_pos_embed_from_grid

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

utils/pos_embed.py:47

↓ 2 callersFunctionget_1d_sincos_pos_embed_from_grid

embed_dim: output dimension for each position pos: a list of positions to be encoded: size (M,) out: (M, D)

Finetuning/util/pos_embed.py:49

↓ 2 callersFunctionget_2d_sincos_pos_embed

grid_size: int of the grid height and width return: pos_embed: [grid_size*grid_size, embed_dim] or [1+grid_size*grid_size, embed_dim] (w/

utils/pos_embed.py:13

↓ 2 callersFunctionget_world_size

()

utils/misc.py:190

↓ 2 callersFunctionget_world_size

()

Finetuning/util/misc.py:195

↓ 2 callersMethodlog_every

(self, iterable, print_freq, header=None)

Finetuning/util/misc.py:123

↓ 2 callersFunctionsave_on_master

(*args, **kwargs)

utils/misc.py:206

↓ 2 callersFunctionsetup_for_distributed

This function disables printing when not in master process

Finetuning/util/misc.py:170

↓ 2 callersMethodstep

(self, hidden_states, conv_state, ssm_state)

mamba_simple.py:452

↓ 2 callersMethodsynchronize_between_processes

(self)

Finetuning/util/misc.py:116

↓ 2 callersMethodupdate

(self, value, n=1)

utils/misc.py:32

↓ 2 callersMethodupdate

(self, value, n=1)

Finetuning/util/misc.py:37

↓ 1 callersMethod__init__

( self, d_model, d_state=16, d_conv=4, expand=2, dt_rank="auto

mamba_simple.py:36

↓ 1 callersMethod__init__

( self, d_model, d_state=16, d_conv=4, expand=2, dt_rank="auto

Finetuning/mamba_simple.py:37

↓ 1 callersMethod_get_states_from_cache

(self, inference_params, batch_size, initialize_states=False)

mamba_simple.py:512

↓ 1 callersMethod_get_states_from_cache

(self, inference_params, batch_size, initialize_states=False)

Finetuning/mamba_simple.py:515

↓ 1 callersMethodadd_meter

(self, name, meter)

utils/misc.py:115

↓ 1 callersMethodadd_meter

(self, name, meter)

Finetuning/util/misc.py:120

↓ 1 callersMethodallocate_inference_cache

(self, batch_size, max_seqlen, dtype=None, **kwargs)

models_pretrain.py:198

↓ 1 callersMethodallocate_inference_cache

(self, batch_size, max_seqlen, dtype=None, **kwargs)

mamba_simple.py:499

↓ 1 callersMethodallocate_inference_cache

(self, batch_size, max_seqlen, dtype=None, **kwargs)

Finetuning/mamba_simple.py:502

↓ 1 callersMethodallocate_inference_cache

(self, batch_size, max_seqlen, dtype=None, **kwargs)

Finetuning/models_mamba.py:342

↓ 1 callersFunctionbuild_transform

(is_train, args)

utils/datasets.py:27

↓ 1 callersFunctionbuild_transform

(is_train, args)

Finetuning/util/datasets.py:31

↓ 1 callersFunctioncreate_block

( d_model, ssm_cfg=None, norm_epsilon=1e-5, drop_path=0., rms_norm=Fal

models_pretrain.py:202

↓ 1 callersFunctioncreate_block

( d_model, ssm_cfg=None, norm_epsilon=1e-5, drop_path=0., rms_norm=False, residual_in_

Finetuning/models_mamba.py:118

↓ 1 callersMethodforward_decoder

(self, latent_ar, decoder_pos_embed)

models_pretrain.py:458

↓ 1 callersMethodforward_features

(self, x, inference_params=None)

models_pretrain.py:423

↓ 1 callersMethodforward_features

(self, x, inference_params=None, if_random_cls_token_position=False, if_random_token_rank=False)

Finetuning/models_mamba.py:356

↓ 1 callersMethodforward_loss

(self, imgs, pred)

models_pretrain.py:491

↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid

(embed_dim, grid)

utils/pos_embed.py:36

↓ 1 callersFunctionget_2d_sincos_pos_embed_from_grid

(embed_dim, grid)

Finetuning/util/pos_embed.py:38

↓ 1 callersFunctionget_args_parser

()

main_pretrain.py:30

↓ 1 callersFunctionget_args_parser

()

Finetuning/main_finetune.py:43

↓ 1 callersFunctionget_grad_norm_

(parameters, norm_type: float = 2.0)

utils/misc.py:265

↓ 1 callersFunctionget_grad_norm_

(parameters, norm_type: float = 2.0)

Finetuning/util/misc.py:280

↓ 1 callersFunctionget_layer_id_for_vit

Assign a parameter with its layer id Following BEiT: https://github.com/microsoft/unilm/blob/master/beit/optim_factory.py#L33

utils/lr_decay.py:65

↓ 1 callersFunctionget_layer_id_for_vit

Assign a parameter with its layer id Following BEiT: https://github.com/microsoft/unilm/blob/master/beit/optim_factory.py#L33

Finetuning/util/lr_decay.py:63

↓ 1 callersFunctionget_rank

()

utils/misc.py:196

↓ 1 callersFunctionget_rank

()

Finetuning/util/misc.py:201

↓ 1 callersFunctioninterpolate_pos_embed

(model, checkpoint_model)

Finetuning/util/pos_embed.py:75

↓ 1 callersFunctionis_main_process

()

utils/misc.py:202

↓ 1 callersFunctionis_main_process

()

Finetuning/util/misc.py:207

↓ 1 callersMethodlog_every

(self, iterable, print_freq, header=None)

utils/misc.py:118

↓ 1 callersFunctionmain

(args)

main_pretrain.py:99

↓ 1 callersFunctionmain

(args)

Finetuning/main_finetune.py:159

↓ 1 callersMethodmask_generate

(self, segment, tokens_per_segment)

models_pretrain.py:393

Finetuning/util/misc.py:70

↓ 1 callersMethodno_weight_decay

(self)

Finetuning/models_mamba.py:349

↓ 1 callersMethodpatchify

imgs: (N, 3, H, W) x: (N, L, patch_size**2 *3)

models_pretrain.py:477

↓ 1 callersFunctionsave_on_master

(*args, **kwargs)

Finetuning/util/misc.py:211

↓ 1 callersFunctionsetup_for_distributed

This function disables printing when not in master process

utils/misc.py:165

↓ 1 callersMethodstep

(self, hidden_states, conv_state, ssm_state)

Finetuning/mamba_simple.py:455

↓ 1 callersMethodstep

(self)

Finetuning/util/lars.py:23

↓ 1 callersMethodsynchronize_between_processes

(self)

utils/misc.py:111

↓ 1 callersFunctiontrain_one_epoch

(model: torch.nn.Module, data_loader: Iterable, optimizer: torch.optim.Optimizer,

engine_pretrain.py:17

↓ 1 callersFunctiontrain_one_epoch

(model: torch.nn.Module, model_ema, criterion: torch.nn.Module, data_loader: Iterable, opt

Finetuning/engine_finetune.py:25

Method__call__

(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True)

utils/misc.py:242

Method__call__

(self, loss, optimizer, clip_grad=None, parameters=None, create_graph=False, update_grad=True)

Finetuning/util/misc.py:257

Finetuning/util/misc.py:100

Method__init__

(self, dim, num_heads=8, qkv_bias=False, qk_scale=None, attn_drop=0., proj_drop=0.)

models_pretrain.py:56

Method__init__

(self, dim, num_heads, mlp_ratio=4., qkv_bias=False, qk_scale=None, drop=0., attn_drop=0., dr

models_pretrain.py:88

Method__init__

(self, img_size=224, patch_size=16, stride=16, in_chans=3, embed_dim=768, norm_layer=None, fl

models_pretrain.py:112

Method__init__

(self, in_features, hidden_features=None, out_features=None, act_layer=nn.SiLU, drop=0., norm

models_pretrain.py:138

Method__init__

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection" This Block has a slightly different structure co

models_pretrain.py:165

Method__init__

(self, img_size=224, patch_size=16, stride=16,

models_pretrain.py:286

Method__init__

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection" This Block has a slightly different structure co

mamba_simple.py:542

Method__init__

(self, window_size=20, fmt=None)

utils/misc.py:24

Method__init__

(self, delimiter="\t")

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection" This Block has a slightly different structure co

Finetuning/mamba_simple.py:545

Method__init__

(self, img_size=224, patch_size=16, stride=16, in_chans=3, embed_dim=768, norm_layer=None, flatten=True)

Finetuning/models_mamba.py:31

Method__init__

(self, in_features, hidden_features=None, out_features=None, act_layer=nn.SiLU, drop=0., norm

Finetuning/models_mamba.py:56

Method__init__

Simple block wrapping a mixer class with LayerNorm/RMSNorm and residual connection" This Block has a slightly different structure co

Finetuning/models_mamba.py:83

Method__init__

(self, params, lr=0, weight_decay=0, momentum=0.9, trust_coefficient=0.001)

Finetuning/util/lars.py:18

Method__init__

(self, window_size=20, fmt=None)

Finetuning/util/misc.py:29

Method__init__

(self, delimiter="\t")

Finetuning/util/misc.py:87

Method__init__

(self)

Finetuning/util/misc.py:254

Finetuning/util/misc.py:77

Method__str__

(self)

Finetuning/util/misc.py:108

Function_init_weights

( module, n_layer, initializer_range=0.02, # Now only used for embedding layer.

models_pretrain.py:238

Function_init_weights

( module, n_layer, initializer_range=0.02, # Now only used for embedding layer. rescale_preno

Finetuning/models_mamba.py:156

next →1–100 of 161, ranked by callers