Adapters¤

Module: generative_models.core.adapters

Source: generative_models/core/adapters.py

Overview¤

Model adapter classes for different architectures.

This module provides standardized interfaces for scaling different model architectures with consistent APIs and optimization strategies.

All implementations follow JAX/Flax NNX best practices and provide hardware-aware optimization for different model types.

class DiffusionAdapter

class EnergyAdapter

class ModelAdapter

class ModelSpecs

class TransformerAdapter

def __init__()

def __init__()

def __init__()

def __init__()

def apply_sharding()

def apply_sharding()

def apply_sharding()

def apply_sharding()

def create_diffusion_adapter()

def create_energy_adapter()

def create_transformer_adapter()

def estimate_memory_usage()

def estimate_memory_usage()

def estimate_memory_usage()

def estimate_memory_usage()

def get_model_specs()

def get_model_specs()

def get_model_specs()

def get_model_specs()

def get_optimal_batch_size()

def get_optimal_batch_size()

def get_optimal_batch_size()

def get_optimal_batch_size()

def get_performance_characteristics()