Index

A

Aesthetics (class in drlx.reward_modelling.aesthetics)

AverageBlueReward (class in drlx.reward_modelling.toy_rewards)

B

BaseConditionalDenoiser (class in drlx.denoisers)

BaseTrainer (class in drlx.trainer)

C

cfg_rescale() (drlx.sampling.Sampler method) compute_loss() (drlx.sampling.DDPOSampler method)

create_loader() (drlx.pipeline.Pipeline method) create_train_loader() (drlx.pipeline.Pipeline method) create_val_loader() (drlx.pipeline.Pipeline method)

D

DDPOSampler (class in drlx.sampling) DDPOTrainer (class in drlx.trainer.ddpo_trainer) decode() (drlx.denoisers.BaseConditionalDenoiser method) drlx.denoisers module drlx.denoisers.ldm_unet module drlx.pipeline.imagenet_animal_prompts module

drlx.reward_modelling.pickscore module drlx.reward_modelling.toy_rewards module drlx.sampling module, [1] drlx.trainer module drlx.trainer.ddpo_trainer module

E

encode() (drlx.denoisers.BaseConditionalDenoiser method)

encode_jpeg() (drlx.reward_modelling.toy_rewards.JPEGCompressability method) extract_pipeline() (drlx.trainer.ddpo_trainer.DDPOTrainer method)

F

forward() (drlx.denoisers.BaseConditionalDenoiser method) (drlx.denoisers.ldm_unet.LDMUNet method) (drlx.reward_modelling.aesthetics.Aesthetics method) (drlx.reward_modelling.toy_rewards.AverageBlueReward method) (drlx.reward_modelling.toy_rewards.JPEGCompressability method)

from_pretrained_pipeline() (drlx.denoisers.ldm_unet.LDMUNet method)

G

get_arch() (drlx.trainer.BaseTrainer method)

get_input_shape() (drlx.denoisers.BaseConditionalDenoiser method) (drlx.denoisers.ldm_unet.LDMUNet method)

I

ImagenetAnimalPrompts (class in drlx.pipeline.imagenet_animal_prompts)

J

JPEGCompressability (class in drlx.reward_modelling.toy_rewards)

L

LDMUNet (class in drlx.denoisers.ldm_unet) load_checkpoint() (drlx.trainer.BaseTrainer method) (drlx.trainer.ddpo_trainer.DDPOTrainer method)

loss() (drlx.trainer.ddpo_trainer.DDPOTrainer method)

M

make_default_collate() (drlx.pipeline.Pipeline class method) (drlx.pipeline.PromptPipeline class method) module drlx.denoisers drlx.denoisers.ldm_unet drlx.pipeline.imagenet_animal_prompts drlx.reward_modelling.pickscore drlx.reward_modelling.toy_rewards drlx.sampling, [1] drlx.trainer drlx.trainer.ddpo_trainer

P

PickScoreModel (class in drlx.reward_modelling.pickscore) Pipeline (class in drlx.pipeline) postprocess() (drlx.denoisers.BaseConditionalDenoiser method) (drlx.denoisers.ldm_unet.LDMUNet method)

preprocess() (drlx.denoisers.BaseConditionalDenoiser method) (drlx.denoisers.ldm_unet.LDMUNet method) (drlx.reward_modelling.pickscore.PickScoreModel method) PromptPipeline (class in drlx.pipeline)

S

sample() (drlx.denoisers.BaseConditionalDenoiser method) (drlx.sampling.DDPOSampler method) (drlx.sampling.Sampler method) (drlx.trainer.ddpo_trainer.DDPOTrainer method) sample_and_calculate_rewards() (drlx.trainer.ddpo_trainer.DDPOTrainer method) Sampler (class in drlx.sampling)

save_checkpoint() (drlx.trainer.BaseTrainer method) (drlx.trainer.ddpo_trainer.DDPOTrainer method) save_pretrained() (drlx.trainer.ddpo_trainer.DDPOTrainer method) setup_model() (drlx.trainer.ddpo_trainer.DDPOTrainer method) setup_optimizer() (drlx.trainer.BaseTrainer method) setup_scheduler() (drlx.trainer.BaseTrainer method) step_and_logprobs() (drlx.sampling.DDPOSampler method)

T

train() (drlx.trainer.BaseTrainer method) (drlx.trainer.ddpo_trainer.DDPOTrainer method) training (drlx.denoisers.BaseConditionalDenoiser attribute) (drlx.denoisers.ldm_unet.LDMUNet attribute) (drlx.reward_modelling.aesthetics.Aesthetics attribute) (drlx.reward_modelling.pickscore.PickScoreModel attribute) (drlx.reward_modelling.toy_rewards.AverageBlueReward attribute) (drlx.reward_modelling.toy_rewards.JPEGCompressability attribute)