MedLatents Documentation
Discrete and continuous latent generative models for medical imagery.
Getting Started
User Guides
API Reference
- Autoregressive
- MaskGIT
- Diffusion API
- Flow Matching API
- Bayesian Flow Networks API
- Networks
DiscreteTransformerContinuousTransformerAttentionAttentionWithValueResidualTransformerBlockprecompute_freqs_cis()apply_rotary_emb()init_weights()RMSNormRMSNormZeroDiscreteDiTContinuousDiTDiTBlockDiTBlockWithCrossAttentionDiTCrossAttentionMMDiTBlockDiTAttentionFinalLayerLabelEmbedderDiscreteSequenceEmbedderContinuousSequenceEmbedderTimestepEmbedderSPRINTConfigSPRINTDiTSPRINTSchedulerTokenSelectorTokenRestorer
- Sampling
sample_nucleus()sample_min_p()sample_autoregressive()sample_with_cfg()TemperatureSchedulerconfidence_based_schedule()adaptive_masking()gumbel_max_sampling()MaskGITSchedulerRunningConfidenceRemaskerRunningConfidenceSchedulerRCRGeneratorKLASSGeneratorDDIMSchedulerDPMSolverSchedulerEulerDiscreteSchedulerAncestralSamplingSchedulersample_with_cfg_flow()sample_with_cfg_diffusion()CFGFlowMatcherCFGDiffusergenerate_with_cfg()rescale_cfg()sample_with_dynamic_cfg()sample_bayesian_flow()sample_with_early_stopping()get_confidence()BayesianFlowSamplerDecoupledSTGumbelSoftmaxReinMaxGuidedSamplerget_guidance_schedule()constant_guidance()linear_guidance()cosine_guidance()cosine_decay_guidance()triangular_guidance()cfg_zero_guidance()cfg_zero_star_guidance()dynamic_guidance()is_compile_available()compile_model()compile_function()CompiledSampler
- Training API
- Generation API
- Inference API
- Post-Training API
- Evaluation API
- Conditioning API
- Rasterization API
- Data API
- Configs API
Tutorials
Research
Project Info
Indices and tables
Features
Core Models
Autoregressive Transformer: Unidirectional generation with causal attention
MaskGIT: Bidirectional masked transformer for parallel decoding
Discrete DiT: Diffusion transformer for discrete latent spaces (D3PM)
Flow Matching: Discrete and continuous flow-based generation
Bayesian Flow Networks: Probabilistic flow models
Advanced Features
Halton Scheduler: Spatially-dispersed unmasking for MaskGIT
KLASS Early Stopping: Adaptive stopping based on KL divergence
Time-Dependent CFG: Dynamic classifier-free guidance schedules
Speculative Decoding: Accelerated autoregressive generation
Rectified Flow++: Iterative reflow for straighter trajectories
Medical Domain
Rasterization: Space-filling curves (Hilbert, Z-order) for spatial-to-sequence
Inpainting: Fill missing regions in medical images
Super-Resolution: Upscale low-resolution medical scans
Clinical Metrics: Dice, IoU, PSNR, SSIM evaluation