+h%SSKJr SSKJrJrJr SSKrSSKrSSKJ r SSKJ r SSK J r Jr SSKJr SS KJr \R&R"S S 55r\"S S \55r"SS\\ 5rg)) dataclass)OptionalTupleUnionN)random) ConfigMixinregister_to_config) BaseOutput)FlaxSchedulerMixinc\rSrSr%Sr\\\S'Sr\\ R\S'Sr \\ R\S'\ S5r Srg)KarrasVeSchedulerStateNnum_inference_steps timestepsschedulecU"5$N)clss h/home/james-whalen/.local/lib/python3.13/site-packages/diffusers/schedulers/scheduling_karras_ve_flax.pycreateKarrasVeSchedulerState.create$s u r)__name__ __module__ __qualname____firstlineno__rrint__annotations__rjnpndarrayr classmethodr__static_attributes__rrrrrsK*.#-'+Ix $+&*Hhs{{#*rrc`\rSrSr%Sr\R \S'\R \S'\\S'Sr g)FlaxKarrasVeOutput)a Output class for the scheduler's step function output. Args: prev_sample (`jnp.ndarray` of shape `(batch_size, num_channels, height, width)` for images): Computed sample (x_{t-1}) of previous timestep. `prev_sample` should be used as next model input in the denoising loop. derivative (`jnp.ndarray` of shape `(batch_size, num_channels, height, width)` for images): Derivative of predicted original image sample (x_0). state (`KarrasVeSchedulerState`): the `FlaxKarrasVeScheduler` state data class. prev_sample derivativestaterN) rrrr__doc__r"r#r!rr%rrrr'r')s#   !!rr'c\rSrSrSr\S5r\S!S\S\S\S\S\S \4 S jj5r S r S"S \ S\ S\ S\ 4SjjrS \ S\R S\S\R$S\ \R \44 SjrS#S \ S\R S\S\S\R S\S\\\ 44SjjrS#S \ S\R S\S\S\R S\R S\R S\S\\\ 44SjjrS \ 4SjrS rg )$FlaxKarrasVeScheduler<a Stochastic sampling from Karras et al. [1] tailored to the Variance-Expanding (VE) models [2]. Use Algorithm 2 and the VE column of Table 1 from [1] for reference. [1] Karras, Tero, et al. "Elucidating the Design Space of Diffusion-Based Generative Models." https://huggingface.co/papers/2206.00364 [2] Song, Yang, et al. "Score-based generative modeling through stochastic differential equations." https://huggingface.co/papers/2011.13456 [`~ConfigMixin`] takes care of storing all config attributes that are passed in the scheduler's `__init__` function, such as `num_train_timesteps`. They can be accessed via `scheduler.config.num_train_timesteps`. [`SchedulerMixin`] provides general loading and saving functionality via the [`SchedulerMixin.save_pretrained`] and [`~SchedulerMixin.from_pretrained`] functions. For more details on the parameters, see the original paper's Appendix E.: "Elucidating the Design Space of Diffusion-Based Generative Models." https://huggingface.co/papers/2206.00364. The grid search values used to find the optimal {s_noise, s_churn, s_min, s_max} for a specific model are described in Table 5 of the paper. Args: sigma_min (`float`): minimum noise magnitude sigma_max (`float`): maximum noise magnitude s_noise (`float`): the amount of additional noise to counteract loss of detail during sampling. A reasonable range is [1.000, 1.011]. s_churn (`float`): the parameter controlling the overall amount of stochasticity. A reasonable range is [0, 100]. s_min (`float`): the start value of the sigma range where we add noise (enable stochasticity). A reasonable range is [0, 10]. s_max (`float`): the end value of the sigma range where we add noise. A reasonable range is [0.2, 80]. cg)NTrselfs r has_stateFlaxKarrasVeScheduler.has_state[sr sigma_min sigma_maxs_noises_churns_mins_maxcgrr)r2r5r6r7r8r9r:s r__init__FlaxKarrasVeScheduler.__init___s rc*[R5$r)rrr1s r create_state"FlaxKarrasVeScheduler.create_stateks%,,..rrr+rshapereturnc[R"SU5SSS2R5nUVs/sHXnURRS-URR S-URRS-- XRS- - --PMZ nnUR U[R"U[RS9US9$s snf)aO Sets the continuous timesteps used for the diffusion chain. Supporting function to be run before inference. Args: state (`KarrasVeSchedulerState`): the `FlaxKarrasVeScheduler` state data class. num_inference_steps (`int`): the number of diffusion steps used when generating samples with a pre-trained model. rNrr )dtype)rrr) r"arangecopyconfigr6r5replacearrayfloat32)r2r+rrArirs r set_timesteps#FlaxKarrasVeScheduler.set_timestepsnsJJq"56tt<AAC    %%q(;;((!+dkk.C.CQ.FFAghQhLijk  }} 3YYxs{{;   sACsamplesigmakeycURRUs=::aURR::a1O O.[URRUR - S5nOSn[ R"USS9nURR[ R"XBRS9-nX5U--nX'S-US-- S-U--nX4$)u Explicit Langevin-like "churn" step of adding noise to the sample according to a factor gamma_i ≥ 0 to reach a higher noise level sigma_hat = sigma_i + gamma_i*sigma_i. TODO Args: g4y?rr )num)rQrAr?) rHr9r:minr8rrsplitr7normalrA) r2r+rOrPrQgammaeps sigma_hat sample_hats radd_noise_to_input(FlaxKarrasVeScheduler.add_noise_to_inputs ;;   :):): : ++e.G.GGTEEll3A&kk!!FMMc$NNEM) 1 uax 7C?#EF $$r model_outputrZ sigma_prevr[ return_dictc\XSU--nXW- U- nXTU- U--n U(dXU4$[XUS9$)a} Predict the sample at the previous timestep by reversing the SDE. Core function to propagate the diffusion process from the learned model outputs (most often the predicted noise). Args: state (`KarrasVeSchedulerState`): the `FlaxKarrasVeScheduler` state data class. model_output (`torch.Tensor` or `np.ndarray`): direct output from learned diffusion model. sigma_hat (`float`): TODO sigma_prev (`float`): TODO sample_hat (`torch.Tensor` or `np.ndarray`): TODO return_dict (`bool`): option for returning tuple rather than FlaxKarrasVeOutput class Returns: [`~schedulers.scheduling_karras_ve_flax.FlaxKarrasVeOutput`] or `tuple`: Updated sample in the diffusion chain and derivative. [`~schedulers.scheduling_karras_ve_flax.FlaxKarrasVeOutput`] if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is the sample tensor. r)r*r+r') r2r+r^rZr_r[r`pred_original_sampler* sample_prevs rstepFlaxKarrasVeScheduler.stepsM6 * ,DD 79D  $:j#HH U3 3!kX]^^rrer*c nXdU--n Xi- U- n XTU- SU-SU ----nU(dXgU4$[XgUS9$)a Correct the predicted sample based on the output model_output of the network. TODO complete description Args: state (`KarrasVeSchedulerState`): the `FlaxKarrasVeScheduler` state data class. model_output (`torch.Tensor` or `np.ndarray`): direct output from learned diffusion model. sigma_hat (`float`): TODO sigma_prev (`float`): TODO sample_hat (`torch.Tensor` or `np.ndarray`): TODO sample_prev (`torch.Tensor` or `np.ndarray`): TODO derivative (`torch.Tensor` or `np.ndarray`): TODO return_dict (`bool`): option for returning tuple rather than FlaxKarrasVeOutput class Returns: prev_sample (TODO): updated sample in the diffusion chain. derivative (TODO): TODO rTrbrc) r2r+r^rZr_r[rer*r`rdderivative_corrs r step_correct"FlaxKarrasVeScheduler.step_corrects^8 +,-FF&=K $:sZ?ORUXgRg?g#hh U3 3!kX]^^rc[5er)NotImplementedError)r2r+original_samplesnoisers r add_noiseFlaxKarrasVeScheduler.add_noises !##rN)g{Gz?dg&1?Pg?2)r)T)rrrrr,propertyr3r floatr<r?rr rrMr"r#jaxArrayr\boolrr'rfrjrpr%rrrr.r.<s<                 /WY + BE NS  8%%% % % YY % s{{E! " %B!"_%"_kk"_ "_  "_ KK "_"_ !5( )"_Z!#_%#_kk#_ #_  #_ KK #_[[#_KK#_#_ !5( )#_J$5$rr.) dataclassesrtypingrrrflaxrw jax.numpynumpyr"rconfiguration_utilsr r utilsr scheduling_utils_flaxr structrr'r.rrrrst ")) A5 "" "$r$. r$r