MCPcopy Create free account

hub / github.com/bilibili/Index-anisora / functions

Functions18,851 in github.com/bilibili/Index-anisora

⨍ Functions 18,851◇ Types & classes 4,392↳ Endpoints 34
↓ 12 callersFunctiondefault
(val, d)
anisora_rl/sat/sgm/modules/attention.py:64
↓ 12 callersFunctiondefault
(val, d)
anisoraV1_train_npu/sgm/modules/attention.py:64
↓ 12 callersFunctiondefault
(val, d)
anisoraV1_train_npu/sat/sgm/modules/attention.py:64
↓ 12 callersFunctiondefault
(val, d)
anisoraV1_infer/fastercache/models/cogvideox/sgm/modules/attention.py:64
↓ 12 callersFunctiondefault
(val, d)
anisoraV1_train_gpu/sgm/modules/attention.py:64
↓ 12 callersFunctiondefault
(val, d)
anisoraV1_train_gpu/sat/sgm/modules/attention.py:64
↓ 12 callersMethodfrom_pretrained
(cls, name="vgg_lpips")
anisora_rl/sat/sgm/modules/autoencoding/lpips/loss/lpips.py:34
↓ 12 callersFunctionget_activation_layer
get activation layer Args: act_type (str): the activation type Returns: torch.nn.functional: the activation layer
anisoraV2_gpu/fastvideo/models/hunyuan/modules/activation_layers.py:4
↓ 12 callersFunctionget_activation_layer
get activation layer Args: act_type (str): the activation type Returns: torch.nn.functional: the activation layer
anisoraV2_npu/fastvideo/models/hunyuan/modules/activation_layers.py:4
↓ 12 callersFunctionget_model_parallel_rank
Return my rank for the model parallel group.
anisora_rl/SwissArmyTransformer-main/sat/mpu/initialize.py:140
↓ 12 callersFunctionget_model_parallel_rank
Return my rank for the model parallel group.
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:141
↓ 12 callersFunctionget_model_parallel_rank
Return my rank for the model parallel group.
anisoraV1_infer/sat/sat/mpu/initialize.py:141
↓ 12 callersFunctionget_model_parallel_rank
Return my rank for the model parallel group.
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:141
↓ 12 callersFunctionget_temporal_pad
()
anisoraV1_infer/fastercache/dsp/comm.py:394
↓ 12 callersFunctionnormalization
Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.
anisoraV1_train_npu/sat/sgm/modules/diffusionmodules/util.py:228
↓ 12 callersFunctionnormalization
Make a standard normalization layer. :param channels: number of input channels. :return: an nn.Module for normalization.
anisoraV1_train_gpu/sat/sgm/modules/diffusionmodules/util.py:228
↓ 12 callersMethodpad
For a list of images, for each images, pads a batch of images to the bottom and right of the image with zeros to the size of largest height a
reward/mantis/models/idefics3/image_processing_idefics3.py:538
↓ 12 callersMethodpreprocess
(self)
anisoraV1_infer/videosys/core/shardformer/t5/policy.py:11
↓ 12 callersFunctionprint_all
(msg, level=logging.INFO, flush=True, log_file=None)
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/helpers.py:159
↓ 12 callersFunctionprint_all
(msg, level=logging.INFO, flush=True, log_file=None)
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/helpers.py:159
↓ 12 callersFunctionsetEffect
( name: keyof Effects, index: number, options?: EffectOptions, )
reward/character/samurai/sam2/demo/frontend/src/common/components/video/Video.tsx:197
↓ 12 callersMethodset_processor
r""" Set the attention processor to use. Args: processor (`AttnProcessor`): The attention processor to us
anisoraV1_infer/fastercache/models/vchitect/attention.py:445
↓ 12 callersFunctionshrink_head
(encoder_state, dim)
anisoraV2_gpu/fastvideo/models/hunyuan/modules/attenion.py:44
↓ 12 callersFunctionshrink_head
(encoder_state, dim)
anisoraV2_npu/fastvideo/models/hunyuan/modules/attenion.py:44
↓ 12 callersMethodt_mask_select
(self, x_mask, x, masked_x, T, S)
anisoraV1_infer/fastercache/models/opensora/stdit3.py:93
↓ 12 callersFunctionuseScreenSize
()
reward/character/samurai/sam2/demo/frontend/src/common/screen/useScreenSize.tsx:19
↓ 11 callersFunctionNormalize
(in_channels, num_groups=32)
anisoraV1_infer/videosys/models/autoencoders/autoencoder_kl_open_sora_plan_v110.py:28
↓ 11 callersMethod__init__
( self, in_channels: int, out_channels: int, factor_t, factor_s=1,
anisoraV3.2/wan/modules/vae2_2.py:372
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisora_rl/sat/sgm/modules/diffusionmodules/openaimodel.py:1225
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisoraV1_train_npu/sgm/modules/diffusionmodules/openaimodel.py:1287
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisoraV1_train_npu/sat/sgm/modules/diffusionmodules/openaimodel.py:1225
↓ 11 callersMethod__init__
(self, config: Idefics3Config)
reward/mantis/models/idefics3/modeling_idefics3.py:817
↓ 11 callersMethod__init__
(self, freq=10000.0, F0=1.0, scaling_factor=1.0)
anisoraV1_infer/videosys/models/transformers/open_sora_plan_v110_transformer_3d.py:137
↓ 11 callersMethod__init__
(self, freq=10000.0, F0=1.0, scaling_factor=1.0)
anisoraV1_infer/fastercache/models/opensora_plan/latte.py:136
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisoraV1_infer/fastercache/models/cogvideox/sgm/modules/diffusionmodules/openaimodel.py:1225
↓ 11 callersMethod__init__
(self)
anisoraV1_infer/fastercache/models/vchitect/attention.py:42
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisoraV1_train_gpu/sgm/modules/diffusionmodules/openaimodel.py:1287
↓ 11 callersMethod__init__
(self, in_channels=3, model_channels=64)
anisoraV1_train_gpu/sat/sgm/modules/diffusionmodules/openaimodel.py:1225
↓ 11 callersMethodadd
(self, val)
reward/character/samurai/sam2/training/utils/train_utils.py:245
↓ 11 callersMethodbatch_to_head_dim
r""" Reshape the tensor from `[batch_size, seq_len, dim]` to `[batch_size // heads, seq_len, dim * heads]`. `heads` is the number of h
anisoraV1_infer/fastercache/models/vchitect/attention.py:607
↓ 11 callersFunctiondecode
(rleObjs: RLEObject[])
reward/character/samurai/sam2/demo/frontend/src/jscocotools/mask.ts:97
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisora_rl/sat/sgm/modules/diffusionmodules/sampling.py:503
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisoraV1_train_npu/sgm/modules/diffusionmodules/sampling.py:551
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisoraV1_train_npu/sat/sgm/modules/diffusionmodules/sampling.py:503
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisoraV1_infer/fastercache/models/cogvideox/sgm/modules/diffusionmodules/sampling.py:503
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisoraV1_train_gpu/sgm/modules/diffusionmodules/sampling.py:551
↓ 11 callersMethoddenoise
(self, x, denoiser, alpha_cumprod_sqrt, cond, uc, timestep=None, idx=None, scale=None, scale_emb=None)
anisoraV1_train_gpu/sat/sgm/modules/diffusionmodules/sampling.py:503
↓ 11 callersFunctiondestroy_model_parallel
Set the groups to none.
anisora_rl/SwissArmyTransformer-main/sat/mpu/initialize.py:176
↓ 11 callersFunctiondestroy_model_parallel
Set the groups to none.
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:177
↓ 11 callersFunctiondestroy_model_parallel
Set the groups to none.
anisoraV1_infer/sat/mpu/initialize.py:177
↓ 11 callersFunctiondestroy_model_parallel
Set the groups to none.
anisoraV1_infer/sat/sat/mpu/initialize.py:177
↓ 11 callersFunctiondestroy_model_parallel
Set the groups to none.
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:177
↓ 11 callersMethoddownload_file
下载远程文件到本地 @param object_name: 远程文件相对路径 @param filename: 本地文件路径
anisoraV2_npu/fastvideo/bili_space/boss/tmp.py:101
↓ 11 callersFunctionenable_pab
()
anisoraV1_infer/fastercache/dsp/pab_mgr.py:131
↓ 11 callersMethodencode
( self, x: torch.Tensor, return_reg_log: bool = False, unregularized: bool = F
anisora_rl/sat/vae_modules/autoencoder41failed.py:219
↓ 11 callersMethodencode
( self, x: torch.Tensor, return_reg_log: bool = False, unregularized: bool = F
anisora_rl/sat/vae_modules/autoencoder.py:219
↓ 11 callersFunctionextract_into_tensor
(a, t, x_shape)
anisoraV2_gpu/fastvideo/distill/solver.py:21
↓ 11 callersFunctionextract_into_tensor
(a, t, x_shape)
anisoraV2_npu/fastvideo/distill/solver.py:21
↓ 11 callersFunctionflash_attention
q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens
anisora_anymask/wan/modules/attention.py:49
↓ 11 callersFunctionflash_attention
q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens
anisoraV3/wan/modules/attention.py:49
↓ 11 callersFunctionflash_attention
q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens
anisoraV2_gpu/wan/modules/attention.py:49
↓ 11 callersFunctionflash_attention
q: [B, Lq, Nq, C1]. k: [B, Lk, Nk, C1]. v: [B, Lk, Nk, C2]. Nq must be divisible by Nk. q_lens
anisoraV2_npu/fastvideo/bili_space/wan/modules/attention.py:30
↓ 11 callersFunctiongather_sequence
(input_, process_group, dim, grad_scale=1.0, pad=0)
anisoraV1_infer/videosys/core/comm.py:362
↓ 11 callersMethodget_object
从BOSS获取远程对象字节到本地内存 @param object_name: 远程对象相对路径 @return 成功返回对象字节数据，否则抛异常
anisoraV2_npu/fastvideo/bili_space/boss/tmp.py:116
↓ 11 callersFunctionget_sequence_parallel_size
()
anisoraV1_train_npu/sat/transformer_sp/util.py:13
↓ 11 callersFunctioninitialize_model_parallel
Initialize model data parallel groups. Arguments: model_parallel_size: number of GPUs used to parallelize model. Let's say we h
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:35
↓ 11 callersFunctioninitialize_model_parallel
Initialize model data parallel groups. Arguments: model_parallel_size: number of GPUs used to parallelize model. Let's say we h
anisoraV1_infer/sat/sat/mpu/initialize.py:35
↓ 11 callersFunctioninitialize_model_parallel
Initialize model data parallel groups. Arguments: model_parallel_size: number of GPUs used to parallelize model. Let's say we h
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/mpu/initialize.py:35
↓ 11 callersMethodis_initialized
(self)
anisoraV1_train_gpu/sat/sgm/dist_group_mgr.py:31
↓ 11 callersFunctionnonlinearity
(x)
anisora_rl/SwissArmyTransformer-main/sat/tokenization/cogview/vqvae/vqvae_diffusion.py:29
↓ 11 callersFunctionnonlinearity
(x)
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/tokenization/cogview/vqvae/vqvae_diffusion.py:29
↓ 11 callersFunctionnonlinearity
(x)
anisoraV1_infer/sat/tokenization/cogview/vqvae/vqvae_diffusion.py:29
↓ 11 callersFunctionnonlinearity
(x)
anisoraV1_infer/sat/sat/tokenization/cogview/vqvae/vqvae_diffusion.py:29
↓ 11 callersFunctionnonlinearity
(x)
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/tokenization/cogview/vqvae/vqvae_diffusion.py:29
↓ 11 callersMethodread_video
(self, video_path, bound=None)
reward/mantis/benchmark/mvbench_eval_utils.py:505
↓ 11 callersMethodsendResponse
( action: T['action'], message?: Omit<T, 'action'>, transfer?: Transferable[], )
reward/character/samurai/sam2/demo/frontend/src/common/components/video/VideoWorkerContext.ts:508
↓ 10 callersMethod__init__
(self, freq=10000.0, F0=1.0, interpolation_scale_thw=(1, 1, 1))
anisoraV1_infer/videosys/models/transformers/open_sora_plan_v120_transformer_3d.py:64
↓ 10 callersMethod__init__
(self, in_features: int, out_features: int, *args, **kwargs)
anisoraV1_infer/fastercache/models/genmo/mochi_preview/vae/models.py:171
↓ 10 callersMethod__setattr__
(self, __name, __value)
anisoraV1_infer/sat/model/registry.py:37
↓ 10 callersFunction_conv_split
(input_, dim, kernel_size)
anisora_rl/sat/vae_modules/cp_enc_dec.py:175
↓ 10 callersMethod_get_object_abs_path
(self, object_name)
anisoraV2_npu/fastvideo/bili_space/boss/__init__.py:68
↓ 10 callersMethod_get_object_abs_path
(self, object_name)
anisoraV2_npu/fastvideo/bili_space/boss/tmp.py:68
↓ 10 callersMethod_sigma_to_alpha_sigma_t
(self, sigma)
anisoraV3.2/wan/utils/fm_solvers.py:335
↓ 10 callersMethod_sigma_to_alpha_sigma_t
(self, sigma)
anisora_anymask/wan/utils/fm_solvers.py:333
↓ 10 callersMethod_sigma_to_alpha_sigma_t
(self, sigma)
anisoraV3/wan/utils/fm_solvers.py:333
↓ 10 callersMethod_sigma_to_alpha_sigma_t
(self, sigma)
anisoraV2_gpu/wan/utils/fm_solvers.py:333
↓ 10 callersMethod_sigma_to_alpha_sigma_t
(self, sigma)
anisoraV2_npu/fastvideo/bili_space/wan/utils/fm_solvers.py:333
↓ 10 callersFunctioncheckpoint
Evaluate a function without caching intermediate activations, allowing for reduced memory at the expense of extra compute in the backward pas
anisoraV1_infer/fastercache/models/cogvideox/vae_modules/utils.py:348
↓ 10 callersMethodclip_grad_norm_
(max_norm)
reward/mantis/models/openflamingo/flamingo.py:294
↓ 10 callersFunctioncopy_layer_param
in-place copy from src to dst src and dst should be the same layer type, e.g., both are LayerNorm or both are Linear. Or at least, bo
anisoraV1_infer/sat/examples/cait/transform_param.py:24
↓ 10 callersFunctioncopy_layer_param
in-place copy from src to dst src and dst should be the same layer type, e.g., both are LayerNorm or both are Linear. Or at least, bo
anisoraV1_infer/sat/examples/mixtral-8x7b/transform_param.py:47
↓ 10 callersFunctioncopy_to_model_parallel_region
(input_)
anisoraV1_infer/sat/mpu/mappings.py:131
↓ 10 callersFunctioncustom
(start, end, kw_args_index, cross_layer_index)
anisoraV1_infer/fastercache_sample_cogvideox_sp.py:265
↓ 10 callersMethoddecode
(self, zs)
anisoraV2_npu/fastvideo/bili_space/wan/modules/vae.py:657
↓ 10 callersMethoddynamic_switch
(self, x, s, t, to_spatial_shard: bool)
anisoraV1_infer/fastercache/models/opensora/stdit3.py:186
↓ 10 callersFunctionfilling_sequence
seq: [2, 3, 5, ..., -1(to be generated), -1, ...] mems: [num_layers, batch_size, len_mems(index), mem_hidden_size] cache,
anisoraV1_infer/sat/generation/autoregressive_sampling.py:52
↓ 10 callersMethodfrom_pretrained
(cls, name, args=None, *, home_path=None, url=None, prefix='', build_only=False, use_node_group=True, overwrit
anisora_rl/SwissArmyTransformer-main/sat/model/base_model.py:215
↓ 10 callersMethodfrom_pretrained
(cls, name, args=None, *, home_path=None, url=None, prefix='', build_only=False, use_node_group=True, overwrit
anisoraV1_train_npu/swissarmytransformer-npu_t_sp/sat/model/base_model.py:214
↓ 10 callersMethodfrom_pretrained
(cls, name, args=None, *, home_path=None, url=None, prefix='', build_only=False, use_node_group=True, overwrit
anisoraV1_train_gpu/swissarmytransformer-npu_t_sp/sat/model/base_model.py:214
↓ 10 callersMethodfrom_pretrained_base
Load a pretrained checkpoint of the current model. Args: name: The identifier of the pretrained model. arg
anisora_rl/SwissArmyTransformer-main/sat/model/base_model.py:187
← previousnext →301–400 of 18,851, ranked by callers