Фреймворки
Libraries that structure an application around a model - orchestration, evaluation and serving. They differ mainly in what they assume about how much control you want.
12 technologies
- PyTorch framework The dominant deep learning framework, built around dynamic computation graphs.
- TensorFlow framework Google's deep learning framework, oriented toward deployment at scale.
- Sentence Transformers framework Library for embedding sentences and passages into vectors.
- Diffusers library Library of diffusion model implementations and pipelines.
- Keras framework High-level modelling API that runs on several backends.
- DSPy framework Builds model pipelines as programs whose prompts are optimised rather than written.
- DeepSpeed framework Training optimisation library for large models across many GPUs.
- AutoGen framework Framework for multi-agent conversations where agents critique each other.
- Unsloth framework Fine-tuning implementation with hand-written kernels for speed and low memory.
- JAX framework NumPy-like array library with automatic differentiation and XLA compilation.
- CrewAI framework Defines agent teams with roles and tasks, and runs them in order.
- Axolotl framework Configuration-driven fine-tuning runner for many model architectures.