MCPcopy Create free account

hub / github.com/NVlabs/Sana / functions

Functions2,606 in github.com/NVlabs/Sana

⨍ Functions 2,606◇ Types & classes 463↳ Endpoints 11
↓ 1 callersFunctionget_args
()
app/app_sana.py:168
↓ 1 callersFunctionget_args
()
scripts/inference_image_reward.py:209
↓ 1 callersFunctionget_args
()
scripts/inference_dpg.py:221
↓ 1 callersFunctionget_args
()
scripts/inference_sana_sprint.py:202
↓ 1 callersFunctionget_args
()
scripts/interface.py:54
↓ 1 callersFunctionget_args
()
scripts/inference.py:228
↓ 1 callersFunctionget_args
()
inference_video_scripts/inference_sana_video.py:341
↓ 1 callersMethodget_by_name
(self, name)
diffusion/data/wids/wids_mmtar.py:118
↓ 1 callersMethodget_cached_modules_by_block
Locate ``CachedCausalAttention`` and ``CachedGLUMBConvTemp`` modules. The result is a list (one entry per transformer block) of the cached
diffusion/scheduler/self_forcing_flow_euler_sampler.py:213
↓ 1 callersMethodget_coefficients_exponential_negative
Calculate the integral of exp(-x) * x^order dx from interval_start to interval_end
diffusion/scheduler/sa_solver_diffusers.py:414
↓ 1 callersMethodget_coefficients_exponential_negative
Calculate the integral of exp(-x) * x^order dx from interval_start to interval_end For calculating the coefficient of gradient terms
diffusion/model/sa_solver.py:463
↓ 1 callersMethodget_coefficients_exponential_positive
Calculate the integral of exp(x(1+tau^2)) * x^order dx from interval_start to interval_end
diffusion/scheduler/sa_solver_diffusers.py:438
↓ 1 callersMethodget_coefficients_exponential_positive
Calculate the integral of exp(x(1+tau^2)) * x^order dx from interval_start to interval_end For calculating the coefficient of gradien
diffusion/model/sa_solver.py:490
↓ 1 callersFunctionget_cosine_decay_to_constant_with_warmup
Create a schedule with a cosine annealing lr followed by a constant lr. Args: optimizer ([`~torch.optim.Optimizer`]): Th
diffusion/utils/lr_scheduler.py:62
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/vk_q_mm_relu_bwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/mm.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/pad_vk_mm_fwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/linear_relu_fwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/vk_q_mm_divide_fwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/vk_mm_relu_bwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_lite_mla_kernels/proj_divide_bwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_mb_conv_pre_glu_kernels/linear_glu_fwd.py:24
↓ 1 callersFunctionget_cuda_autotune_config
()
diffusion/model/nets/fastlinear/modules/triton_mb_conv_pre_glu_kernels/depthwise_conv_fwd.py:24
↓ 1 callersMethodget_current_decay
(self, optimization_step)
diffusion/post_training/ema.py:23
↓ 1 callersFunctionget_denoising_loss
(loss_type: str)
diffusion/longsana/utils/loss.py:107
↓ 1 callersFunctionget_dist_local_rank
()
diffusion/model/dc_ae/efficientvit/apps/utils/dist.py:71
↓ 1 callersFunctionget_dist_size
()
diffusion/model/dc_ae/efficientvit/apps/utils/dist.py:58
↓ 1 callersMethodget_file
(self, i)
diffusion/data/wids/wids_tar.py:89
↓ 1 callersFunctionget_files
(idx, base_dir)
tools/inference_scaling/nvila_sana_pick.py:66
↓ 1 callersMethodget_head_inputs
(self)
diffusion/model/nets/sana_ladd.py:43
↓ 1 callersFunctionget_header_str
()
app/app_sana_4bit_compare_bf16.py:239
↓ 1 callersFunctionget_image_encoder
(name, model_path, tokenizer_path=None, device="cuda", dtype=None, config=None)
diffusion/model/builder.py:104
↓ 1 callersFunctionget_local_rank
()
diffusion/utils/dist_utils.py:52
↓ 1 callersMethodget_mask_latent
rescale the mask to latent space size according to latent blended diffusion
app/sana_pipeline_inpaint.py:281
↓ 1 callersMethodget_orders_and_timesteps_for_singlestep_solver
Get the order of each step for sampling by the singlestep DPM-Solver. We combine both DPM-Solver-1,2,3 to use all the function evalu
diffusion/model/dpm_solver.py:891
↓ 1 callersFunctionget_pipeline
( precision: str, use_qencoder: bool = False, device: str | torch.device = "cuda", pipeline_init_kwargs: d
app/app_sana_4bit_compare_bf16.py:100
↓ 1 callersFunctionget_pose_inverse
Computes the inverse of a batch of 4x4 homogeneous transformation matrices T using the R^T = R^-1 property for rotation matrices. T: (...
diffusion/utils/cam_utils.py:121
↓ 1 callersFunctionget_prompt
(idx, base_dir)
tools/inference_scaling/nvila_sana_pick.py:57
↓ 1 callersFunctionget_rmsnorm_class
Return the RMSNorm class to use, controlled by ``SANA_USE_LIGER`` env var. Liger is **on by default** when the package is installed. Set ``SA
diffusion/model/liger_norms.py:36
↓ 1 callersMethodget_rope
(self, pos_embed_type, attention_head_dim, patch_size, rope_fhw_dim)
diffusion/model/nets/sana_multi_scale_video.py:486
↓ 1 callersMethodget_scalings_for_boundary_condition_discrete
(self, t)
diffusion/scheduler/lcm_scheduler.py:310
↓ 1 callersMethodget_shard
(self, url)
diffusion/data/wids/wids.py:418
↓ 1 callersMethodget_shard
Get the shard and index within the shard corresponding to the given index.
diffusion/data/wids/wids.py:606
↓ 1 callersFunctionget_sigmas
(timesteps, n_dim=4, dtype=torch.float32)
train_scripts/train_dreambooth_lora_sana.py:1300
↓ 1 callersFunctionget_tl_dtype_from_torch_dtype
(dtype: torch.dtype)
diffusion/model/nets/fastlinear/modules/utils/dtype.py:32
↓ 1 callersFunctionget_version
(rel_path)
tools/metrics/pytorch-fid/setup.py:12
↓ 1 callersFunctionget_version
(rel_path)
tools/metrics/clip-score/setup.py:12
↓ 1 callersMethodgetdata
(self, idx)
diffusion/data/datasets/sana_data.py:93
↓ 1 callersMethodgetdata
(self, idx)
diffusion/data/datasets/sana_data.py:321
↓ 1 callersMethodgetdata
(self, idx)
diffusion/data/datasets/sana_data_multi_scale.py:121
↓ 1 callersMethodgetdata
(self, idx)
diffusion/data/datasets/video/sana_video_data.py:284
↓ 1 callersFunctiongroup_by_key
Group the file names by key. Args: names: A list of file names. Returns: A list of lists of indices, where each sublist cont
diffusion/data/wids/wids.py:122
↓ 1 callersFunctionguidance_type_select
(default_guidance_type, pag_scale, attn_type)
tools/controlnet/inference_controlnet.py:99
↓ 1 callersFunctionhf_download_or_fpath
Backwards-compatible alias for :func:`resolve_hf_path`.
sana/tools/hf_utils.py:59
↓ 1 callersFunctionimages_are_same
Check if two PIL images are the same
app/app_sana_inpaint.py:418
↓ 1 callersMethodinference_with_trajectory
denoise in chunk-wise manner, input/output is consistent with sample in SelfForcingFlowEuler.sample: - input/output: latents (B, C, T
diffusion/longsana/pipeline/sana_training_pipeline.py:340
↓ 1 callersMethodinit_cam_branch_from_base
(self)
diffusion/model/nets/sana_multi_scale_video_camctrl.py:1527
↓ 1 callersMethodinit_cam_branch_weights
Copy main-branch QKV weights into the camera branch for transfer learning.
diffusion/model/nets/sana_gdn_camctrl_blocks.py:293
↓ 1 callersFunctioninit_modules
(model: Union[nn.Module, list[nn.Module]], init_type="trunc_normal")
diffusion/model/dc_ae/efficientvit/apps/utils/init.py:26
↓ 1 callersMethodinit_weights
r""" Initialize model parameters using Xavier initialization.
diffusion/model/wan/model.py:971
↓ 1 callersMethodinitialize
(self)
diffusion/model/nets/sana_multi_scale_adaln.py:334
↓ 1 callersMethodinitialize
(self)
diffusion/model/nets/sana_multi_scale_video_camctrl.py:1263
↓ 1 callersMethodinitialize
(self)
diffusion/model/nets/sana_multi_scale_video.py:803
↓ 1 callersMethodinitialize
(self)
diffusion/model/nets/sana_U_shape_multi_scale.py:326
↓ 1 callersMethodinitialize
(self)
diffusion/model/nets/sana_multi_scale.py:423
↓ 1 callersMethodinitialize_all
(self)
diffusion/model/nets/sana_multi_scale_controlnet.py:160
↓ 1 callersMethodinitialize_all_and_copy_from_base
(self, base_block)
diffusion/model/nets/sana_multi_scale_controlnet.py:53
↓ 1 callersMethodinitialize_weights
(self)
diffusion/model/nets/sana_U_shape.py:314
↓ 1 callersFunctionintersect_range
Return the intersection of the two half-open integer intervals.
diffusion/data/wids/wids.py:811
↓ 1 callersMethodinverse
Inverse the sample `x` from time `t_start` to `t_end` by DPM-Solver. For discrete-time DPMs, we use `t_start=1/N`, where `N` is the t
diffusion/model/dpm_solver.py:1504
↓ 1 callersMethodinverse_lambda
Compute the continuous-time label t in [0, T] of a given half-logSNR lambda_t.
diffusion/model/dpm_solver.py:250
↓ 1 callersFunctionis_connector_key
(key: str)
tools/convert_sana_wm_refiner_to_diffusers.py:181
↓ 1 callersFunctionis_local_master
()
diffusion/utils/dist_utils.py:65
↓ 1 callersFunctionis_transformer_key
(key: str)
tools/convert_sana_wm_refiner_to_diffusers.py:129
↓ 1 callersMethodlagrange_polynomial_coefficient
Calculate the coefficient of lagrange polynomial
diffusion/scheduler/sa_solver_diffusers.py:482
↓ 1 callersMethodlagrange_polynomial_coefficient
Calculate the coefficient of lagrange polynomial For lagrange interpolation
diffusion/model/sa_solver.py:537
↓ 1 callersFunctionlengths_to_ranges
Convert a list of lengths to a list of ranges.
diffusion/data/wids/wids.py:801
↓ 1 callersFunctionlinear_glu_fwd
Input: x: (..., C) weight: (2*D, C) bias: (2*D,) Output: y: (..., D)
diffusion/model/nets/fastlinear/modules/triton_mb_conv_pre_glu_kernels/linear_glu_fwd.py:203
↓ 1 callersFunctionlist_mean
(x: list)
diffusion/model/dc_ae/efficientvit/models/utils/list.py:34
↓ 1 callersFunctionload_checkpoint_ddp
( checkpoint, model, model_ema=None, optimizer=None, lr_scheduler=None, load_ema=False
diffusion/utils/checkpoint.py:251
↓ 1 callersFunctionload_checkpoint_fsdp
( checkpoint, model, remove_state_dict_keys=None, )
diffusion/utils/checkpoint.py:334
↓ 1 callersFunctionload_image_from_dir
( dir_path: str, suffix: Union[str, tuple[str, ...], list[str]] = (".jpg", ".JPEG", ".png"), retur
diffusion/model/dc_ae/efficientvit/apps/utils/image.py:36
↓ 1 callersFunctionload_jsonl
(file_path: str)
scripts/inference_geneval.py:78
↓ 1 callersFunctionload_jsonl
(file_path: str)
scripts/inference_sana_sprint_geneval.py:77
↓ 1 callersFunctionload_model
(model_name)
tools/inference_scaling/nvila_sana_pick.py:15
↓ 1 callersFunctionload_models
(args)
tools/metrics/geneval/evaluation/evaluate_images.py:49
↓ 1 callersMethodload_state_dict
(self, state_dict: dict)
diffusion/post_training/ema.py:79
↓ 1 callersFunctionlog_results
(df_dict)
tools/metrics/geneval/evaluation/evaluate_images.py:341
↓ 1 callersFunctionmain
()
tools/convert_sana_wm_refiner_to_diffusers.py:253
↓ 1 callersFunctionmain
()
tools/metrics/dpg_bench/compute_dpg_bench.py:187
↓ 1 callersFunctionmain
()
tools/metrics/pytorch-fid/compute_fid.py:203
↓ 1 callersFunctionmain
()
tools/metrics/pytorch-fid/src/pytorch_fid/fid_score.py:278
↓ 1 callersFunctionmain
(opt)
tools/metrics/geneval/generation/diffusers_generate.py:84
↓ 1 callersFunctionmain
(args)
tools/metrics/geneval/evaluation/evaluate_images.py:250
↓ 1 callersFunctionmain
()
tools/metrics/clip-score/clip_score.py:262
↓ 1 callersFunctionmain
()
tools/metrics/clip-score/src/clip_score/clip_score.py:175
↓ 1 callersFunctionmain
()
tools/metrics/image_reward/compute_image_reward.py:55
↓ 1 callersFunctionmain
(args)
tools/convert_scripts/convert_sana_video_to_diffusers.py:33
↓ 1 callersFunctionmain
(args)
tools/convert_scripts/convert_sana_to_diffusers.py:47
↓ 1 callersFunctionmain
(args)
tools/convert_scripts/convert_sana_to_svdquant.py:47
← previousnext →1,101–1,200 of 2,606, ranked by callers