Functions — github.com/DreamEditBenchTeam/DreamEdit

hub / github.com/DreamEditBenchTeam/DreamEdit / functions

Functions598 in github.com/DreamEditBenchTeam/DreamEdit

⨍ Functions 598◇ Types & classes 104

↓ 2 callersFunctionNormalize

(in_channels)

src/ldm/modules/attention.py:76

↓ 2 callersMethod_get_denoise_row_from_list

(self, samples, desc='', force_no_decoder_quantization=False)

src/ldm/models/diffusion/ddpm.py:530

↓ 2 callersMethod_get_rows_from_list

(self, samples)

src/ldm/models/diffusion/ddpm.py:370

↓ 2 callersMethod_validation_step

(self, batch, batch_idx, suffix="")

src/ldm/models/autoencoder.py:170

↓ 2 callersFunctionadd_Gaussian_noise

(img, noise_level1=2, noise_level2=25)

src/ldm/modules/image_degradation/bsrgan_light.py:373

↓ 2 callersFunctionadd_Poisson_noise

(img)

src/ldm/modules/image_degradation/bsrgan.py:404

↓ 2 callersFunctionadd_resize

(img, sf=4)

src/ldm/modules/image_degradation/bsrgan.py:339

↓ 2 callersFunctionadd_speckle_noise

(img, noise_level1=2, noise_level2=25)

src/ldm/modules/image_degradation/bsrgan.py:386

↓ 2 callersFunctionalways

(val)

src/ldm/modules/x_transformer.py:64

↓ 2 callersFunctionbicubic_degradation

Args: x: HxWxC image, [0, 1] sf: down-scale factor Return: bicubicly downsampled LR image

src/ldm/modules/image_degradation/bsrgan_light.py:228

↓ 2 callersFunctionbicubic_degradation

Args: x: HxWxC image, [0, 1] sf: down-scale factor Return: bicubicly downsampled LR image

src/ldm/modules/image_degradation/bsrgan.py:228

↓ 2 callersFunctioncompute_cosine_distance

(image_features, image_features2)

src/metrics/distances.py:4

↓ 2 callersMethodcopy_to

(self, model)

src/ldm/modules/ema.py:46

↓ 2 callersFunctioncount_flops_attn

A counter for the `thop` package to count the operations in an attention operation. Meant to be used like: macs, params = thop.pr

src/ldm/modules/diffusionmodules/openaimodel.py:327

↓ 2 callersFunctioncount_params

(model, verbose=False)

src/ldm/util.py:71

↓ 2 callersFunctioncubic

(x)

src/ldm/modules/image_degradation/utils_image.py:700

↓ 2 callersMethoddata_prediction_fn

Return the data prediction model (with thresholding).

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:386

↓ 2 callersMethoddecode

(self, quant)

src/ldm/models/autoencoder.py:107

↓ 2 callersMethoddecode

(self, z)

src/ldm/models/autoencoder.py:330

↓ 2 callersMethoddelta_border

:param h: height :param w: width :return: normalized distance to image border, wtith min distance = 0 at border and

src/ldm/models/diffusion/ddpm.py:571

↓ 2 callersFunctiondream_edit

( src_img_path, device, config, obj_img_path=None, bbox_file_path=None

src/iterate_generate.py:451

↓ 2 callersMethodencode_text

(self, prompt)

src/metrics/clip_vit.py:21

↓ 2 callersFunctionexists

(x)

src/ldm/util.py:53

↓ 2 callersMethodfind_in_interval

(self, n)

src/ldm/lr_scheduler.py:52

↓ 2 callersFunctiongeo_mean

(iterable)

src/auto_evaluation.py:10

↓ 2 callersMethodget_conditioning

(self, batch, k=None)

src/ldm/models/diffusion/classifier.py:133

↓ 2 callersMethodget_input

(self, batch, k)

src/ldm/models/diffusion/classifier.py:124

↓ 2 callersMethodget_input

(self, batch, k, return_first_stage_outputs=False, force_c_encode=False, cond_key=None, retu

src/ldm/models/diffusion/ddpm.py:654

↓ 2 callersFunctionget_model_input_time

Convert the continuous-time `t_continuous` (in [epsilon, T]) to the model input time. For discrete-time DPMs, we convert `t_continuou

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:278

↓ 2 callersFunctionget_polished_mask

param: mask in size [H, W] k = kernel number mask_type name of mask type return: mask in size [H, W]

src/utils/mask_helper.py:34

↓ 2 callersFunctiongroup_dict_by_key

(cond, d)

src/ldm/modules/x_transformer.py:93

↓ 2 callersFunctiongroupby_prefix_and_trim

(prefix, d)

src/ldm/modules/x_transformer.py:110

↓ 2 callersMethodinit_from_ckpt

(self, path, ignore_keys=list(), only_model=False)

src/ldm/models/diffusion/ddpm.py:186

↓ 2 callersFunctioninterpolate_fn

A piecewise linear function y = f(x), using xp and yp as keypoints. We implement f(x) in a differentiable way (i.e. applicable for autograd).

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:1132

↓ 2 callersFunctionismap

(x)

src/ldm/util.py:41

↓ 2 callersFunctioniterative_edit

( src_img_path, # background image img_to_edit_path, obj_class, lang_sam_mode

src/iterate_generate.py:106

↓ 2 callersFunctionlatent_to_image

(model, latents)

src/generate_new.py:92

↓ 2 callersFunctionload_model_from_config

(config, ckpt, verbose=False)

src/generate_new.py:47

↓ 2 callersFunctionmake_ddim_sampling_parameters

(alphacums, ddim_timesteps, eta, verbose=True)

src/ldm/modules/diffusionmodules/util.py:63

↓ 2 callersFunctionmake_ddim_timesteps

(ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, verbose=True)

src/ldm/modules/diffusionmodules/util.py:46

↓ 2 callersFunctionmkdir

(path)

src/ldm/modules/image_degradation/utils_image.py:153

↓ 2 callersMethodnoise_prediction_fn

Return the noise prediction model.

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:380

↓ 2 callersMethodp_sample

(self, x, c, t, clip_denoised=False, repeat_noise=False, return_codebook_ids=False, quantize_

src/ldm/models/diffusion/ddpm.py:1079

↓ 2 callersMethodp_sample_ddim

(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,

src/ldm/models/diffusion/ddim.py:166

↓ 2 callersFunctionpil_to_tensor

param: pil_img - Image Object return: tensor in size [1, C, H, W]

src/utils/visual_helper.py:50

↓ 2 callersMethodpredict_start_from_noise

(self, x_t, t, noise)

src/ldm/models/diffusion/ddpm.py:216

↓ 2 callersMethodq_posterior

(self, x_start, x_t, t)

src/ldm/models/diffusion/ddpm.py:222

↓ 2 callersFunctionrandom_crop

(lq, hq, sf=4, lq_patchsize=64)

src/ldm/modules/image_degradation/bsrgan.py:427

↓ 2 callersMethodregister_schedule

(self, given_betas=None, beta_schedule="linear", timesteps=1000, linear_start=1e-4,

src/ldm/models/diffusion/ddpm.py:117

↓ 2 callersMethodrestore

Restore the parameters stored with the `store` method. Useful to validate the model with EMA parameters without affecting the

src/ldm/modules/ema.py:64

↓ 2 callersFunctionshift_pixel

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

src/ldm/modules/image_degradation/bsrgan_light.py:99

↓ 2 callersFunctionshift_pixel

shift pixel for super-resolution with different scale factors Args: x: WxHxC or WxH sf: scale factor upper_left: shift dir

src/ldm/modules/image_degradation/bsrgan.py:99

↓ 2 callersMethodsinglestep_dpm_solver_third_update

Singlestep solver DPM-Solver-3 from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at time `s`.

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:633

↓ 2 callersMethodstore

Save the current parameters for restoring later. Args: parameters: Iterable of `torch.nn.Parameter`; the parameters to be

src/ldm/modules/ema.py:55

↓ 2 callersFunctionsubtract_mask

small_mask must be $\in$ big_mask param: big_mask in size [H, W] small_mask in size [H, W] return: mask in size [

src/utils/mask_helper.py:20

↓ 2 callersFunctiontensor_to_pil

param: tensor_img in size [1, C, H, W] return: pil - Image Object

src/utils/visual_helper.py:61

↓ 2 callersFunctiontimestep_embedding

Create sinusoidal timestep embeddings. :param timesteps: a 1-D Tensor of N indices, one per batch element. These may be

src/ldm/modules/diffusionmodules/util.py:151

↓ 2 callersMethodto_rgb

(self, x)

src/ldm/models/autoencoder.py:417

↓ 2 callersFunctiontransform_box_mask_paste

(labeled_box, sam_box, mask, background_image, subject_image)

src/utils/mask_helper.py:113

↓ 1 callersMethod__len__

(self)

src/ldm/data/base.py:18

↓ 1 callersFunction_augment

(img)

src/ldm/modules/image_degradation/utils_image.py:475

↓ 1 callersMethod_filter_relpaths

(self, relpaths)

src/ldm/data/imagenet.py:48

↓ 1 callersFunction_get_paths_from_images

(path)

src/ldm/modules/image_degradation/utils_image.py:74

↓ 1 callersMethod_load

(self)

src/ldm/data/imagenet.py:93

↓ 1 callersMethod_prepare

(self)

src/ldm/data/imagenet.py:45

↓ 1 callersMethod_prepare_human_to_integer_label

(self)

src/ldm/data/imagenet.py:80

↓ 1 callersMethod_prepare_idx_to_synset

(self)

src/ldm/data/imagenet.py:74

↓ 1 callersMethod_prepare_synset_to_human

(self)

src/ldm/data/imagenet.py:66

↓ 1 callersFunctionadd_sharpening

USM sharpening. borrowed from real-ESRGAN Input image: I; Blurry image: B. 1. K = I + weight * (I - B) 2. Mask = 1 if abs(I - B) > thresho

src/ldm/modules/image_degradation/bsrgan.py:299

↓ 1 callersFunctionaddition

( src_img_path, obj_img_path, # Useful only for copy-paste obj_class, lang_sa

src/iterate_generate.py:271

↓ 1 callersFunctionanisotropic_Gaussian

generate an anisotropic Gaussian kernel Args: ksize : e.g., 15, kernel size theta : [0, pi], rotation angle range l1

src/ldm/modules/image_degradation/bsrgan_light.py:65

↓ 1 callersFunctionanisotropic_Gaussian

generate an anisotropic Gaussian kernel Args: ksize : e.g., 15, kernel size theta : [0, pi], rotation angle range l1

src/ldm/modules/image_degradation/bsrgan.py:65

↓ 1 callersFunctionaugment_img

Kai Zhang (github: https://github.com/cszn)

src/ldm/modules/image_degradation/utils_image.py:380

↓ 1 callersFunctionavg_pool_nd

Create a 1D, 2D, or 3D average pooling module.

src/ldm/modules/diffusionmodules/util.py:238

↓ 1 callersFunctionbgr2ycbcr

bgr version of rgb2ycbcr only_y: only return Y channel Input: uint8, [0, 255] float, [0, 1]

src/ldm/modules/image_degradation/utils_image.py:573

↓ 1 callersMethodcalculate_adaptive_weight

(self, nll_loss, g_loss, last_layer=None)

src/ldm/modules/losses/vqperceptual.py:85

↓ 1 callersMethodcalculate_adaptive_weight

(self, nll_loss, g_loss, last_layer=None)

src/ldm/modules/losses/contperceptual.py:32

↓ 1 callersFunctioncond_grad_fn

Compute the gradient of the classifier, i.e. nabla_{x} log p_t(cond | x_t).

src/ldm/models/diffusion/dpm_solver/dpm_solver.py:312

↓ 1 callersMethodddim_sampling

(self, cond, shape, x_T=None, ddim_use_original_steps=False, callb

src/ldm/models/diffusion/ddim.py:114

↓ 1 callersFunctiondiffedit

:param init_image: image to be edit :param src_prompt: prompt describe origin image(i.e. A bowl of fruits) :param dst_prompt: prompt desc

src/generate_new.py:441

↓ 1 callersFunctionedit_all_images_from_class

(config)

src/iterate_generate.py:588

↓ 1 callersFunctionedit_single_img_from_class

(config)

src/iterate_generate.py:676

↓ 1 callersMethodencode

(self, text)

src/ldm/modules/encoders/modules.py:101

↓ 1 callersMethodencode

(self, x)

src/ldm/models/autoencoder.py:324

↓ 1 callersMethodencode_with_pretrained

(self,x)

src/ldm/modules/diffusionmodules/model.py:816

↓ 1 callersFunctionequals

(val)

src/ldm/modules/x_transformer.py:76

↓ 1 callersFunctionevaluate_clipi_score

(real_image, generated_image, device, clip_model)

src/metrics/evaluate_dino.py:29

↓ 1 callersFunctionevaluate_dino_score

(real_image, generated_image, device, fidelity)

src/metrics/evaluate_dino.py:5

↓ 1 callersMethodfreeze

(self)

src/ldm/modules/encoders/modules.py:147

↓ 1 callersFunctionfspecial_gaussian

(hsize, sigma)

src/ldm/modules/image_degradation/bsrgan_light.py:187

↓ 1 callersFunctionfspecial_gaussian

(hsize, sigma)

src/ldm/modules/image_degradation/bsrgan.py:187

↓ 1 callersFunctionfspecial_laplacian

(alpha)

src/ldm/modules/image_degradation/bsrgan_light.py:201

↓ 1 callersFunctionfspecial_laplacian

(alpha)

src/ldm/modules/image_degradation/bsrgan.py:201

↓ 1 callersMethodget_base

(self)

src/ldm/data/imagenet.py:379

↓ 1 callersFunctionget_gligen_inpaint_pipeline

(device)

src/pipelines/inpainting_pipelines.py:47

↓ 1 callersFunctionget_image_paths

(dataroot)

src/ldm/modules/image_degradation/utils_image.py:67

↓ 1 callersFunctionget_mask

the map value will be clamped to map.mean() * clamp_rate, then values will be scaled into 0~1, then term into binary(split at 0.5). so if

src/generate_new.py:108

↓ 1 callersFunctionget_obj_from_str

(string, reload=False)

src/ldm/util.py:88

↓ 1 callersFunctionget_object_caption

(image, class_name, BLIP_Model, langSAM)

src/pipelines/extract_object_pipeline.py:10

↓ 1 callersMethodget_object_caption

(self, image, class_name: str)

src/pipelines/extract_object_pipeline.py:28

← previous next →101–200 of 598, ranked by callers