Latte: Cross-framework Python Package for Evaluation of Latent-Based Generative Models

TL;DR

Latte is a versatile Python library that enables consistent evaluation of latent-based generative models across different deep learning frameworks like PyTorch and TensorFlow, promoting reproducibility and extensibility.

Contribution

It introduces a framework-agnostic, modular library for evaluating latent-based generative models, supporting multiple deep learning frameworks with reproducible metrics.

Findings

Supports both PyTorch and TensorFlow/Keras

Ensures reproducible and deterministic evaluations

Provides extensible APIs for future framework support

Abstract

Latte (for LATent Tensor Evaluation) is a Python library for evaluation of latent-based generative models in the fields of disentanglement learning and controllable generation. Latte is compatible with both PyTorch and TensorFlow/Keras, and provides both functional and modular APIs that can be easily extended to support other deep learning frameworks. Using NumPy-based and framework-agnostic implementation, Latte ensures reproducible, consistent, and deterministic metric calculations regardless of the deep learning framework of choice.