magyarcég
Javítás és szerviz
Keras
  • Javítás és szerviz

Keras

Human-centric deep learning API supporting multiple backends.

  • Provision of Keras 3 and Keras 2 API documentation
  • Hyperparameter tuning and access to pretrained models
  • Guides on model building using Functional API and subclassing
  • Guides on training, evaluation, and custom training loops
  • Provision of code examples for deep learning use cases

Bemutatkozás

Keras is a deep learning API that prioritizes developer experience through smaller, more readable, and maintainable codebases. It operates with a multi-backend approach, allowing seamless work across JAX, TensorFlow, and PyTorch. The platform offers extensive documentation, including developer guides for advanced topics like subclassing layers and custom training loops, and code examples for various use cases in computer vision, NLP, and generative AI. Keras is utilized by major scientific organizations such as CERN, NASA, and NIH, and powers applications like Waymo's self-driving vehicles. The KerasHub library provides Keras 3 implementations of popular model architectures with pretrained checkpoints.

Szolgáltatások, tevékenységek

  • Providing Keras 3 API documentation
  • Providing Keras 2 API documentation
  • Hyperparameter tuning via KerasTuner
  • Access to pretrained models via KerasHub
  • Guides on Functional API usage
  • Guides on Sequential model creation
  • Guides on making new layers and models via subclassing
  • Guides on training and evaluation with built-in methods
  • Guides on customizing fit() with JAX
  • Guides on customizing fit() with TensorFlow
  • Guides on customizing fit() with PyTorch
  • Guides on writing a custom training loop in JAX
  • Guides on writing a custom training loop in TensorFlow
  • Guides on writing a custom training loop in PyTorch
  • Guides on serialization & saving
  • Guides on customizing saving & serialization
  • Guides on writing your own callbacks
  • Guides on transfer learning & fine-tuning
  • Guides on distributed training with JAX
  • Guides on distributed training with TensorFlow
  • Guides on distributed training with PyTorch
  • Guides on distributed training with Keras 3
  • Guides on migrating Keras 2 code to Keras 3
  • Guides on using Keras with NNX backend
  • Guides on Orbax Checkpointing in Keras
  • Guides on Quantization in Keras
  • Guides on 8-bit integer quantization in Keras
  • Guides on 4-bit integer quantization in Keras
  • Guides on GPTQ quantization in Keras
  • Guides on writing quantization-compatible layers in Keras
  • Guides on customizing quantization in Keras
  • Guides on defining a Custom TPU/GPU Kernel
  • Code examples for computer vision tasks
  • Code examples for natural language processing tasks
  • Code examples for generative deep learning tasks
  • Access to GEMMA model documentation
  • Access to LLAMA model documentation
  • Access to STABLE DIFFUSION model documentation
  • Access to MISTRAL model documentation