Functions — github.com/ChenWu98/cycle-diffusion

hub / github.com/ChenWu98/cycle-diffusion / functions

Functions1,414 in github.com/ChenWu98/cycle-diffusion

⨍ Functions 1,414◇ Types & classes 263

↓ 2 callersFunctionbicubic_degradation

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

model/lib/stable_diffusion/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

model/lib/stable_diffusion/ldm/modules/image_degradation/bsrgan.py:228

↓ 2 callersFunctioncalculate_psnr

(img1, img2)

evaluation/utils.py:60

↓ 2 callersFunctioncalculate_ssim

calculate SSIM the same outputs as MATLAB's img1, img2: [0, 255]

evaluation/utils.py:13

↓ 2 callersFunctioncheckpoint

Evaluate a function without caching intermediate activations, allowing for reduced memory at the expense of extra compute in the backward pas

model/lib/ddpm_ddim/models/improved_ddpm/nn.py:124

↓ 2 callersFunctionconfigure

If comm is provided, average all numerical stats across that comm

model/lib/ddpm_ddim/models/improved_ddpm/logger.py:398

↓ 2 callersMethodcopy_to

(self, model)

model/lib/latentdiff/ldm/modules/ema.py:46

↓ 2 callersMethodcopy_to

(self, model)

model/lib/stable_diffusion/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

model/lib/latentdiff/ldm/modules/diffusionmodules/openaimodel.py:327

↓ 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

model/lib/stable_diffusion/ldm/modules/diffusionmodules/openaimodel.py:327

↓ 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

model/lib/ddpm_ddim/models/improved_ddpm/unet.py:313

↓ 2 callersFunctioncount_params

(model, verbose=False)

model/lib/stable_diffusion/ldm/util.py:71

↓ 2 callersFunctioncreate_model

( image_size, num_channels, num_res_blocks, channel_mult="", learn_sigma=False, class_

model/lib/ddpm_ddim/models/improved_ddpm/script_util.py:45

↓ 2 callersFunctioncubic

(x)

model/lib/latentdiff/ldm/modules/image_degradation/utils_image.py:700

↓ 2 callersFunctioncubic

(x)

model/lib/stable_diffusion/ldm/modules/image_degradation/utils_image.py:700

↓ 2 callersFunctioncustom_to_np

(x)

model/lib/latentdiff/sample_diffusion.py:27

↓ 2 callersMethoddecode

(self, z)

model/lib/latentdiff/ldm/models/autoencoder.py:330

↓ 2 callersMethoddecode

(self, quant)

model/lib/stable_diffusion/ldm/models/autoencoder.py:107

↓ 2 callersMethoddecode

(self, z)

model/lib/stable_diffusion/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

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:565

↓ 2 callersMethoddelta_border

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

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:565

↓ 2 callersFunctiondenoising_step

(xt, t, t_next, *, models, logvars, b,

model/lib/ddpm_ddim/utils/diffusion_utils.py:23

↓ 2 callersFunctiondistributed_concat

(tensor: Union[Tuple, List, torch.tensor], num_total_examples: Optional[int] = None)

trainer/trainer.py:43

↓ 2 callersMethodencode_first_stage

(self, x)

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:817

↓ 2 callersMethodevaluate

Run evaluation and returns metrics. The calling script will be responsible for providing a method to compute metrics, as they are ta

trainer/trainer.py:1017

↓ 2 callersFunctionexists

(x)

model/lib/latentdiff/ldm/util.py:46

↓ 2 callersFunctionexists

(x)

model/lib/stable_diffusion/ldm/util.py:53

↓ 2 callersMethodfind_in_interval

(self, n)

model/lib/latentdiff/ldm/lr_scheduler.py:52

↓ 2 callersMethodfind_in_interval

(self, n)

model/lib/stable_diffusion/ldm/lr_scheduler.py:52

↓ 2 callersFunctionget_condition

(model, text, bs)

model/gan_wrapper/latentdiff_stochastic_text_wrapper.py:30

↓ 2 callersFunctionget_condition

(model, text, bs)

model/gan_wrapper/stable_diffusion_stochastic_text_wrapper.py:28

↓ 2 callersMethodget_conditioning

(self, batch, k=None)

model/lib/latentdiff/ldm/models/diffusion/classifier.py:130

↓ 2 callersMethodget_conditioning

(self, batch, k=None)

model/lib/stable_diffusion/ldm/models/diffusion/classifier.py:130

↓ 2 callersFunctionget_evaluator

(evaluator_program)

utils/program_utils.py:12

↓ 2 callersMethodget_first_stage_encoding

(self, encoder_posterior)

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:536

↓ 2 callersMethodget_input

(self, batch, k)

model/lib/latentdiff/ldm/models/diffusion/classifier.py:122

↓ 2 callersMethodget_input

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

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:647

↓ 2 callersMethodget_input

(self, batch, k)

model/lib/stable_diffusion/ldm/models/diffusion/classifier.py:122

↓ 2 callersMethodget_input

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

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:647

↓ 2 callersFunctionget_param_groups_and_shapes

(named_model_params)

model/lib/ddpm_ddim/models/improved_ddpm/fp16_util.py:82

↓ 2 callersFunctionget_parser

()

model/lib/latentdiff/sample_diffusion.py:159

↓ 2 callersFunctionget_parser

(**parser_kwargs)

model/lib/latentdiff/main.py:24

↓ 2 callersFunctiongroup_dict_by_key

(cond, d)

model/lib/latentdiff/ldm/modules/x_transformer.py:93

↓ 2 callersFunctiongroup_dict_by_key

(cond, d)

model/lib/stable_diffusion/ldm/modules/x_transformer.py:93

↓ 2 callersFunctiongroupby_prefix_and_trim

(prefix, d)

model/lib/latentdiff/ldm/modules/x_transformer.py:110

↓ 2 callersFunctiongroupby_prefix_and_trim

(prefix, d)

model/lib/stable_diffusion/ldm/modules/x_transformer.py:110

↓ 2 callersMethodinit_from_ckpt

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

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:185

↓ 2 callersMethodinit_from_ckpt

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

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:186

↓ 2 callersMethodlog_img

(self, pl_module, batch, batch_idx, split="train")

model/lib/latentdiff/main.py:340

↓ 2 callersMethodlog_img

(self, pl_module, batch, batch_idx, split="train")

model/lib/stable_diffusion/main.py:340

↓ 2 callersFunctionmake_ddim_sampling_parameters

(alphacums, ddim_timesteps, eta, verbose=True)

model/lib/latentdiff/ldm/modules/diffusionmodules/util.py:64

↓ 2 callersFunctionmake_ddim_sampling_parameters

(alphacums, ddim_timesteps, eta, verbose=True)

model/lib/stable_diffusion/ldm/modules/diffusionmodules/util.py:64

↓ 2 callersFunctionmake_ddim_timesteps

(ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, verbose=True)

model/lib/latentdiff/ldm/modules/diffusionmodules/util.py:46

↓ 2 callersFunctionmake_ddim_timesteps

(ddim_discr_method, num_ddim_timesteps, num_ddpm_timesteps, verbose=True)

model/lib/stable_diffusion/ldm/modules/diffusionmodules/util.py:46

↓ 2 callersFunctionmake_master_params

Copy model parameters into a (differently-shaped) list of full-precision parameters.

model/lib/ddpm_ddim/models/improved_ddpm/fp16_util.py:35

↓ 2 callersFunctionmkdir

(path)

model/lib/latentdiff/ldm/modules/image_degradation/utils_image.py:153

↓ 2 callersFunctionmkdir

(path)

model/lib/stable_diffusion/ldm/modules/image_degradation/utils_image.py:153

↓ 2 callersFunctionnested_cpu

CPU `tensors` (even if it's a nested list/tuple/dict of tensors).

trainer/trainer.py:87

↓ 2 callersMethodp_sample

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

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:1069

↓ 2 callersMethodp_sample

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

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:1069

↓ 2 callersMethodp_sample_ddim

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

model/lib/latentdiff/ldm/models/diffusion/ddim.py:501

↓ 2 callersMethodp_sample_loop

(self, shape, return_intermediates=False)

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:250

↓ 2 callersFunctionpil_loader

(path)

utils/file_utils.py:29

↓ 2 callersMethodpredict_start_from_noise

(self, x_t, t, noise)

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:215

↓ 2 callersMethodpredict_start_from_noise

(self, x_t, t, noise)

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:216

↓ 2 callersMethodprogressive_denoising

(self, cond, shape, verbose=True, callback=None, quantize_denoised=False, img_ca

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:1099

↓ 2 callersMethodq_posterior

(self, x_start, x_t, t)

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:221

↓ 2 callersMethodq_posterior

(self, x_start, x_t, t)

model/lib/stable_diffusion/ldm/models/diffusion/ddpm.py:222

↓ 2 callersFunctionrandom_crop

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

model/lib/latentdiff/ldm/modules/image_degradation/bsrgan.py:427

↓ 2 callersFunctionrandom_crop

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

model/lib/stable_diffusion/ldm/modules/image_degradation/bsrgan.py:427

↓ 2 callersMethodrefine

(self, S, refine_steps, batch_size, shape,

model/lib/latentdiff/ldm/models/diffusion/ddim.py:112

↓ 2 callersMethodregister_schedule

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

model/lib/latentdiff/ldm/models/diffusion/ddpm.py:117

↓ 2 callersMethodregister_schedule

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

model/lib/stable_diffusion/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

model/lib/latentdiff/ldm/modules/ema.py:64

↓ 2 callersMethodrestore

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

model/lib/stable_diffusion/ldm/modules/ema.py:64

↓ 2 callersFunctionsample_xt

(x0, t, b)

model/gan_wrapper/ddpm_ddim_wrapper.py:310

↓ 2 callersFunctionsave_images

(images: torch.Tensor, output_dir: str, file_prefix: str, nrows: int, iteration: int)

utils/file_utils.py:9

↓ 2 callersMethodsave_model

Will save the model, so you can reload it using :obj:`from_pretrained()`. Will only save from the main process.

trainer/trainer.py:376

↓ 2 callersFunctionshift_pixel

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

model/lib/latentdiff/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

model/lib/latentdiff/ldm/modules/image_degradation/bsrgan.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

model/lib/stable_diffusion/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

model/lib/stable_diffusion/ldm/modules/image_degradation/bsrgan.py:99

↓ 2 callersMethodstore

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

model/lib/latentdiff/ldm/modules/ema.py:55

↓ 2 callersMethodstore

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

model/lib/stable_diffusion/ldm/modules/ema.py:55

↓ 2 callersFunctiontimestep_embedding

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

model/lib/latentdiff/ldm/modules/diffusionmodules/util.py:152

↓ 2 callersFunctiontimestep_embedding

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

model/lib/stable_diffusion/ldm/modules/diffusionmodules/util.py:152

↓ 2 callersMethodto_rgb

(self, x)

model/lib/latentdiff/ldm/models/autoencoder.py:416

↓ 2 callersMethodto_rgb

(self, x)

model/lib/stable_diffusion/ldm/models/autoencoder.py:416

↓ 2 callersFunctionunflatten_master_params

(param_group, master_param)

model/lib/ddpm_ddim/models/improved_ddpm/fp16_util.py:78

↓ 2 callersMethodzero_grad

(self)

model/lib/ddpm_ddim/models/improved_ddpm/fp16_util.py:173

↓ 2 callersFunctionzero_master_grads

(master_params)

model/lib/ddpm_ddim/models/improved_ddpm/fp16_util.py:128

↓ 1 callersMethod__len__

(self)

model/lib/latentdiff/ldm/data/base.py:18

↓ 1 callersMethod__len__

(self)

model/lib/stable_diffusion/ldm/data/base.py:18

↓ 1 callersFunction_augment

(img)

model/lib/latentdiff/ldm/modules/image_degradation/utils_image.py:475

↓ 1 callersFunction_augment

(img)

model/lib/stable_diffusion/ldm/modules/image_degradation/utils_image.py:475

↓ 1 callersFunction_configure_default_logger

()

model/lib/ddpm_ddim/models/improved_ddpm/logger.py:430

↓ 1 callersMethod_ddpm_ddim_encoding

(self, cond, shape, ddim_use_original_steps=False, cal

model/lib/latentdiff/ldm/models/diffusion/ddim.py:448

↓ 1 callersMethod_ddpm_ddim_encoding

(self, cond, shape, ddim_use_original_steps=False, cal

model/lib/stable_diffusion/ldm/models/diffusion/ddim.py:450

↓ 1 callersMethod_do_log

(self, args)

model/lib/ddpm_ddim/models/improved_ddpm/logger.py:353

↓ 1 callersMethod_filter_relpaths

(self, relpaths)

model/lib/latentdiff/ldm/data/imagenet.py:48

← previous next →201–300 of 1,414, ranked by callers