MCPcopy Create free account

hub / github.com/LuChengTHU/dpm-solver / functions

Functions1,626 in github.com/LuChengTHU/dpm-solver

⨍ Functions 1,626◇ Types & classes 286
↓ 1 callersFunctionmodel_fn
The noise predicition model function that is used for DPM-Solver.
dpm_solver_pytorch.py:309
↓ 1 callersFunctionmodel_fn
The noise predicition model function that is used for DPM-Solver.
examples/stable-diffusion/ldm/models/diffusion/dpm_solver/dpm_solver.py:317
↓ 1 callersFunctionmodel_grads_to_master_grads
Copy the gradients from the model parameters into the master parameters from make_master_params().
examples/ddpm_and_guided-diffusion/models/guided_diffusion/fp16_util.py:52
↓ 1 callersFunctionmodel_wrapper
Create a wrapper function for the noise prediction model. DPM-Solver needs to solve the continuous-time diffusion ODEs. For DPMs trained on discr
examples/ddpm_and_guided-diffusion/dpm_solver/sampler.py:170
↓ 1 callersFunctionmpi_weighted_mean
Copied from: https://github.com/openai/baselines/blob/ea25b9e8b234e6ee1bca43083f8f3cf974143998/baselines/common/mpi_util.py#L110 Perform a we
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:412
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
dpm_solver_jax.py:745
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
dpm_solver_pytorch.py:796
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/ddpm_and_guided-diffusion/dpm_solver/sampler.py:796
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/score_sde_pytorch/dpm_solver.py:804
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/stable-diffusion/ldm/models/diffusion/dpm_solver/dpm_solver.py:804
↓ 1 callersMethodmultistep_dpm_solver_second_update
Multistep solver DPM-Solver-2 from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
examples/score_sde_jax/dpm_solver.py:745
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
dpm_solver_jax.py:802
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
dpm_solver_pytorch.py:854
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/ddpm_and_guided-diffusion/dpm_solver/sampler.py:854
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/score_sde_pytorch/dpm_solver.py:862
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A pytorch tensor. The initial value at t
examples/stable-diffusion/ldm/models/diffusion/dpm_solver/dpm_solver.py:862
↓ 1 callersMethodmultistep_dpm_solver_third_update
Multistep solver DPM-Solver-3 from time `t_prev_list[-1]` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
examples/score_sde_jax/dpm_solver.py:802
↓ 1 callersFunctionnondefault_trainer_args
(opt)
examples/stable-diffusion/main.py:126
↓ 1 callersFunctionnormal_kl
source: https://github.com/openai/guided-diffusion/blob/27c20a8fab9cb472df5d6bdd6c8d11c8f430b924/guided_diffusion/losses.py#L12 Compute the K
examples/stable-diffusion/ldm/modules/distributions/distributions.py:65
↓ 1 callersFunctionnot_equals
(val)
examples/stable-diffusion/ldm/modules/x_transformer.py:70
↓ 1 callersMethodnumerical_clip_alpha
For some beta schedules such as cosine schedule, the log-SNR has numerical isssues. We clip the log-SNR near t=T within -5.1 to ensu
dpm_solver_pytorch.py:114
↓ 1 callersMethodnumerical_clip_alpha
For some beta schedules such as cosine schedule, the log-SNR has numerical isssues. We clip the log-SNR near t=T within -5.1 to ensu
examples/ddpm_and_guided-diffusion/dpm_solver/sampler.py:114
↓ 1 callersFunctionnumpy_to_pil
Convert a numpy image or a batch of images to a PIL image.
examples/stable-diffusion/scripts/txt2img.py:37
↓ 1 callersFunctionoptimize_fn
Optimizes with warmup and gradient clipping (disabled if negative).
examples/score_sde_pytorch/losses.py:41
↓ 1 callersFunctionoptimize_fn
Optimizes with warmup and gradient clipping (disabled if negative).
examples/score_sde_jax/losses.py:43
↓ 1 callersFunctionp_div_fn
Pmapped divergence of the drift function.
examples/score_sde_jax/likelihood.py:68
↓ 1 callersMethodp_losses
(self, x_start, t, noise=None)
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:294
↓ 1 callersMethodp_losses
(self, x_start, cond, t, noise=None)
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:1012
↓ 1 callersMethodp_mean_variance
(self, x, t, clip_denoised: bool)
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:231
↓ 1 callersMethodp_mean_variance
(self, x, c, t, clip_denoised: bool, return_codebook_ids=False, quantize_denoised=False,
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:1047
↓ 1 callersMethodp_sample
(self, x, t, clip_denoised=True, repeat_noise=False)
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:244
↓ 1 callersMethodp_sample_loop
(self, cond, shape, return_intermediates=False, x_T=None, verbose=True, callback=None, t
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:1166
↓ 1 callersMethodp_sample_plms
(self, x, c, t, index, repeat_noise=False, use_original_steps=False, quantize_denoised=False,
examples/stable-diffusion/ldm/models/diffusion/plms.py:173
↓ 1 callersFunctionparam_grad_or_zeros
(param)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/fp16_util.py:141
↓ 1 callersFunctionparse_args_and_config
()
examples/ddpm_and_guided-diffusion/main.py:19
↓ 1 callersFunctionpatches_from_image
(img, p_size=512, p_overlap=64, p_max=800)
examples/stable-diffusion/ldm/modules/image_degradation/utils_image.py:93
↓ 1 callersMethodplms_sampling
(self, cond, shape, x_T=None, ddim_use_original_steps=False, callb
examples/stable-diffusion/ldm/models/diffusion/plms.py:115
↓ 1 callersMethodprepare_data
(self)
examples/stable-diffusion/main.py:185
↓ 1 callersFunctionpreprocess
(data)
examples/score_sde_jax/models/utils.py:296
↓ 1 callersMethodpreprocess
(self, x)
examples/stable-diffusion/ldm/modules/encoders/modules.py:216
↓ 1 callersMethodprior_logp
Compute log-density of the prior distribution. Useful for computing the log-likelihood via probability flow ODE. Args: z: latent code
examples/score_sde_pytorch/sde_lib.py:40
↓ 1 callersFunctionprofile_kv
(scopename)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:294
↓ 1 callersMethodq_mean_variance
Get the distribution q(x_t | x_0). :param x_start: the [N x C x ...] tensor of noiseless inputs. :param t: the number of diff
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:204
↓ 1 callersFunctionrandom_crop
(lq, hq, sf=4, lq_patchsize=64)
examples/stable-diffusion/ldm/modules/image_degradation/bsrgan_light.py:431
↓ 1 callersMethodregister_buffer
(self, name, attr)
examples/stable-diffusion/ldm/models/diffusion/dpm_solver/sampler.py:16
↓ 1 callersMethodreset_noise_accs
(self)
examples/stable-diffusion/ldm/models/diffusion/classifier.py:202
↓ 1 callersFunctionresize_op
(img)
examples/score_sde_pytorch/datasets.py:99
↓ 1 callersFunctionresize_op
(img)
examples/score_sde_jax/datasets.py:107
↓ 1 callersFunctionrun
(model, logdir, batch_size=50, vanilla=False, custom_steps=None, eta=None, n_samples=50000, nplog=None)
examples/stable-diffusion/scripts/sample_diffusion.py:108
↓ 1 callersFunctionrun_inception_jit
Running the inception network. Assuming input is within [0, 255].
examples/score_sde_pytorch/evaluation.py:90
↓ 1 callersFunctionrun_inception_jit
Running the inception network. Assuming input is within [0, 255].
examples/score_sde_jax/evaluation.py:90
↓ 1 callersMethodsample
Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. =========================================
dpm_solver_pytorch.py:1047
↓ 1 callersMethodsample
(self)
examples/ddpm_and_guided-diffusion/runners/diffusion.py:253
↓ 1 callersMethodsample
(self, cond, batch_size=16, return_intermediates=False, x_T=None, verbose=True, timesteps=None,
examples/stable-diffusion/ldm/models/diffusion/ddpm.py:1217
↓ 1 callersMethodsample
Compute the sample at time `t_end` by DPM-Solver, given the initial `x` at time `t_start`. =========================================
examples/score_sde_jax/dpm_solver.py:966
↓ 1 callersMethodsample_fid
(self, model, classifier=None)
examples/ddpm_and_guided-diffusion/runners/diffusion.py:407
↓ 1 callersFunctionsave_logs
(logs, path, n_saved=0, key="sample", np_path=None)
examples/stable-diffusion/scripts/sample_diffusion.py:143
↓ 1 callersMethodschedule
(self, n, **kwargs)
examples/stable-diffusion/ldm/lr_scheduler.py:17
↓ 1 callersMethodschedule
(self, n, **kwargs)
examples/stable-diffusion/ldm/lr_scheduler.py:59
↓ 1 callersMethodsearch
(self, x, k)
examples/stable-diffusion/scripts/knn2img.py:135
↓ 1 callersFunctionsearch_ah
(searcher, dims_per_block, aiq_threshold, reorder_k)
examples/stable-diffusion/scripts/train_searcher.py:24
↓ 1 callersFunctionsearch_bruteforce
(searcher)
examples/stable-diffusion/scripts/train_searcher.py:12
↓ 1 callersFunctionsearch_partioned_ah
(searcher, dims_per_block, aiq_threshold, reorder_k, partioning_trainsize, num_leaves,
examples/stable-diffusion/scripts/train_searcher.py:16
↓ 1 callersMethodset_comm
(self, comm)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:385
↓ 1 callersMethodset_level
(self, level)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:382
↓ 1 callersMethodsetup
(self, stage=None)
examples/stable-diffusion/main.py:189
↓ 1 callersFunctionsigmoid
(x)
examples/ddpm_and_guided-diffusion/runners/diffusion.py:82
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
dpm_solver_jax.py:849
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at
dpm_solver_pytorch.py:906
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at
examples/ddpm_and_guided-diffusion/dpm_solver/sampler.py:906
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at
examples/score_sde_pytorch/dpm_solver.py:914
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A pytorch tensor. The initial value at
examples/stable-diffusion/ldm/models/diffusion/dpm_solver/dpm_solver.py:914
↓ 1 callersMethodsinglestep_dpm_solver_update
Singlestep DPM-Solver with the order `order` from time `s` to time `t`. Args: x: A jnp.DeviceArray. The initial value at
examples/score_sde_jax/dpm_solver.py:849
↓ 1 callersFunctionstate_dict_to_master_params
(model, state_dict, use_fp16)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/fp16_util.py:116
↓ 1 callersMethodstochastic_encode
(self, x0, t, use_original_steps=False, noise=None)
examples/stable-diffusion/ldm/models/diffusion/ddim.py:207
↓ 1 callersFunctionsynset2idx
(path_to_yaml="data/index_synset.yaml")
examples/stable-diffusion/ldm/data/imagenet.py:20
↓ 1 callersFunctionto_sparse_list
(l)
dpm_solver_jax.py:1181
↓ 1 callersFunctionto_sparse_list
(l)
examples/score_sde_jax/dpm_solver.py:1181
↓ 1 callersFunctiontrain_searcher
(opt, metric='dot_product', partioning_trainsize=None,
examples/stable-diffusion/scripts/train_searcher.py:62
↓ 1 callersMethodtrain_searcher
(self, k, metric='dot_product', searcher_savedir=None)
examples/stable-diffusion/scripts/knn2img.py:75
↓ 1 callersFunctionunflatten_master_params
Unflatten the master parameters to look like model_params.
examples/ddpm_and_guided-diffusion/models/improved_ddpm/fp16_util.py:64
↓ 1 callersMethodupdate_fn
(self, x, t)
examples/score_sde_pytorch/sampling.py:212
↓ 1 callersMethodupdate_fn
(self, x, t)
examples/score_sde_pytorch/sampling.py:266
↓ 1 callersMethodupdate_fn
(self, x, t)
examples/score_sde_pytorch/sampling.py:346
↓ 1 callersMethodupdate_fn
(self, rng, x, t)
examples/score_sde_jax/sampling.py:219
↓ 1 callersFunctionupfirdn2d_native
( input, kernel, up_x, up_y, down_x, down_y, pad_x0, pad_x1, pad_y0, pad_y1 )
examples/score_sde_pytorch/op/upfirdn2d.py:159
↓ 1 callersFunctionupfirdn_2d
Pad, upsample, FIR filter, and downsample a batch of 2D images. Accepts a batch of 2D images of the shape `[majorDim, inH, inW, minorDim]` an
examples/score_sde_jax/models/up_or_down_sampling.py:212
↓ 1 callersFunctionupsample_conv_2d
Fused `upsample_2d()` followed by `tf.nn.conv2d()`. Padding is performed only once at the beginning, not between the operations. The f
examples/score_sde_pytorch/models/up_or_down_sampling.py:72
↓ 1 callersFunctionupsample_conv_2d
Fused `upsample_2d()` followed by `tf.nn.conv2d()`. Padding is performed only once at the beginning, not between the operations. The f
examples/score_sde_jax/models/up_or_down_sampling.py:89
↓ 1 callersFunctionvariance_scaling
Ported from JAX.
examples/score_sde_pytorch/models/layers.py:54
↓ 1 callersMethodvesde_update_fn
(self, x, t)
examples/score_sde_pytorch/sampling.py:230
↓ 1 callersMethodvesde_update_fn
(self, rng, x, t)
examples/score_sde_jax/sampling.py:237
↓ 1 callersFunctionvisualize_cond_img
(path)
examples/stable-diffusion/notebook_helpers.py:127
↓ 1 callersMethodvpsde_update_fn
(self, x, t)
examples/score_sde_pytorch/sampling.py:242
↓ 1 callersMethodvpsde_update_fn
(self, rng, x, t)
examples/score_sde_jax/sampling.py:249
↓ 1 callersMethodwrite_logs
(self, loss, logits, targets)
examples/stable-diffusion/ldm/models/diffusion/classifier.py:162
↓ 1 callersMethodwritekvs
(self, kvs)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:27
↓ 1 callersMethodwriteseq
(self, seq)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/logger.py:32
↓ 1 callersFunctionzero_grad
(model_params)
examples/ddpm_and_guided-diffusion/models/guided_diffusion/fp16_util.py:133
↓ 1 callersFunctionzero_module
Zero out the parameters of a module and return it.
examples/stable-diffusion/ldm/modules/attention.py:67
← previousnext →601–700 of 1,626, ranked by callers