docs: make Chinese README the default

This commit is contained in:
wehub-resource-sync
2026-07-13 10:09:45 +00:00
parent 4d6942a991
commit 73d6fa01e2
+57 -59
View File
@@ -1,77 +1,80 @@
# Keras 3: Deep Learning for Humans
<!-- WEHUB_ZH_README -->
> [!NOTE]
> 本文档由 WeHub 基于上游 README 翻译整理,属于社区翻译,非官方中文文档。
> [English](./README.en.md) · [原始项目](https://github.com/keras-team/keras) · [上游 README](https://github.com/keras-team/keras/blob/HEAD/README.md)
> 原作者、版权与许可证归属以原始项目及本仓库 LICENSE 文件为准。
Keras 3 is a multi-backend deep learning framework, with support for JAX, TensorFlow, PyTorch, and OpenVINO (for inference-only).
Effortlessly build and train models for computer vision, natural language processing, audio processing,
timeseries forecasting, recommender systems, etc.
# Keras 3:面向人类的深度学习
- **Accelerated model development**: Ship deep learning solutions faster thanks to the high-level UX of Keras
and the availability of easy-to-debug runtimes like PyTorch or JAX eager execution.
- **State-of-the-art performance**: By picking the backend that is the fastest for your model architecture (often JAX!),
leverage speedups ranging from 20% to 350% compared to other frameworks. [Benchmark here](https://keras.io/getting_started/benchmarks/).
- **Datacenter-scale training**: Scale confidently from your laptop to large clusters of GPUs or TPUs.
Keras 3 是一个多后端深度学习框架,支持 JAX、TensorFlow、PyTorch 和 OpenVINO(仅用于推理)。
可轻松构建并训练用于计算机视觉、自然语言处理、音频处理、
时间序列预测、推荐系统等领域的模型。
Join nearly three million developers, from burgeoning startups to global enterprises, in harnessing the power of Keras 3.
- **加速模型开发**:借助 Keras 的高级用户体验,以及 PyTorch 或 JAX 急切执行(eager execution)等易于调试的运行时,更快交付深度学习解决方案。
- **业界领先的性能**:选择最适合您模型架构的后端(通常是 JAX!),相比其他框架可获得 20% 至 350% 的加速。[基准测试在这里](https://keras.io/getting_started/benchmarks/).
- **数据中心级训练**:自信地从笔记本电脑扩展到大型 GPU 或 TPU 集群。
与近三百万开发者一起,从新兴初创公司到全球企业,共同发挥 Keras 3 的威力。
## Installation
## 安装
### Install with pip
### 使用 pip 安装
Keras 3 is available on PyPI as `keras`. Note that Keras 2 remains available as the `tf-keras` package.
Keras 3 PyPI 上提供为 `keras`。请注意,Keras 2 仍以 `tf-keras` 包的形式提供。
1. Install `keras`:
1. 安装 `keras`
```
pip install keras --upgrade
```
2. Install backend package(s).
2. 安装后端包。
To use `keras`, you should also install the backend of choice: `tensorflow`, `jax`, or `torch`. Additionally,
The `openvino` backend is available with support for model inference only.
要使用 `keras`,您还需要安装所选的后端:`tensorflow``jax` `torch`。此外,
`openvino` 后端可用,但仅支持模型推理。
### Local installation
### 本地安装
#### Minimal installation
#### 最小安装
Keras 3 is compatible with Linux and macOS systems. For Windows users, we recommend using WSL2 to run Keras.
To install a local development version:
Keras 3 兼容 Linux macOS 系统。对于 Windows 用户,我们建议使用 WSL2 运行 Keras
要安装本地开发版本:
1. Install dependencies:
1. 安装依赖:
```
pip install -r requirements.txt
```
2. Run installation command from the root directory.
2. 从根目录运行安装命令。
```
python pip_build.py --install
```
3. Run API generation script when creating PRs that update `keras_export` public APIs:
3. 在创建更新 `keras_export` 公共 API 的 PR 时,运行 API 生成脚本:
```
./shell/api_gen.sh
```
## Backend Compatibility Table
## 后端兼容性表
The following table lists the minimum supported versions of each backend for the latest stable release of Keras (v3.x):
下表列出了最新稳定版 Kerasv3.x)对各后端的最低支持版本:
| Backend | Minimum Supported Version |
| 后端 | 最低支持版本 |
|------------|---------------------------|
| TensorFlow | 2.16.1 |
| JAX | 0.4.20 |
| PyTorch | 2.1.0 |
| OpenVINO | 2025.3.0 |
#### Adding GPU support
#### 添加 GPU 支持
The `requirements.txt` file will install a CPU-only version of TensorFlow, JAX, and PyTorch. For GPU support, we also
provide a separate `requirements-{backend}-cuda.txt` for TensorFlow, JAX, and PyTorch. These install all CUDA
dependencies via `pip` and expect a NVIDIA driver to be pre-installed. We recommend a clean Python environment for each
backend to avoid CUDA version mismatches. As an example, here is how to create a JAX GPU environment with `conda`:
`requirements.txt` 文件将安装仅 CPU 版本的 TensorFlowJAX PyTorch。如需 GPU 支持,我们还
为 TensorFlow、JAX 和 PyTorch 提供了单独的 `requirements-{backend}-cuda.txt`。这些会通过 `pip` 安装所有 CUDA
依赖,并期望已预装 NVIDIA 驱动。我们建议为每个后端使用干净的 Python 环境,以避免 CUDA 版本不匹配。例如,以下是如何使用 `conda` 创建 JAX GPU 环境:
```shell
conda create -y -n keras-jax python=3.10
@@ -80,16 +83,16 @@ pip install -r requirements-jax-cuda.txt
python pip_build.py --install
```
## Configuring your backend
## 配置后端
You can export the environment variable `KERAS_BACKEND` or you can edit your local config file at `~/.keras/keras.json`
to configure your backend. Available backend options are: `"tensorflow"`, `"jax"`, `"torch"`, `"openvino"`. Example:
您可以导出环境变量 `KERAS_BACKEND`,或编辑本地配置文件 `~/.keras/keras.json`
来配置后端。可用的后端选项有:`"tensorflow"``"jax"``"torch"``"openvino"`。示例:
```
export KERAS_BACKEND="jax"
```
In Colab, you can do:
Colab 中,您可以:
```python
import os
@@ -98,36 +101,31 @@ os.environ["KERAS_BACKEND"] = "jax"
import keras
```
**Note:** The backend must be configured before importing `keras`, and the backend cannot be changed after
the package has been imported.
**注意:** 必须在导入 `keras` 之前配置后端,导入包后无法更改后端。
**Note:** The OpenVINO backend is an inference-only backend, meaning it is designed only for running model
predictions using `model.predict()` method.
**注意:** OpenVINO 后端是仅推理后端,意味着它仅设计用于通过 `model.predict()` 方法运行模型预测。
## Backwards compatibility
## 向后兼容
Keras 3 is intended to work as a drop-in replacement for `tf.keras` (when using the TensorFlow backend). Just take your
existing `tf.keras` code, make sure that your calls to `model.save()` are using the up-to-date `.keras` format, and you're
done.
Keras 3 旨在作为 `tf.keras` 的直接替代(使用 TensorFlow 后端时)。只需拿出现有的 `tf.keras` 代码,确保对 `model.save()` 的调用使用最新的 `.keras` 格式,即可完成。
If your `tf.keras` model does not include custom components, you can start running it on top of JAX or PyTorch immediately.
如果您的 `tf.keras` 模型不包含自定义组件,您可以立即在 JAX PyTorch 上运行它。
If it does include custom components (e.g. custom layers or a custom `train_step()`), it is usually possible to convert it
to a backend-agnostic implementation in just a few minutes.
如果确实包含自定义组件(例如自定义层或自定义 `train_step()`),通常只需几分钟即可将其转换为与后端无关的实现。
In addition, Keras models can consume datasets in any format, regardless of the backend you're using:
you can train your models with your existing `tf.data.Dataset` pipelines or PyTorch `DataLoaders`.
此外,无论使用哪个后端,Keras 模型都可以使用任何格式的数据集:
您可以使用现有的 `tf.data.Dataset` 流水线或 PyTorch `DataLoaders` 训练模型。
## Why use Keras 3?
## 为何使用 Keras 3
- Run your high-level Keras workflows on top of any framework -- benefiting at will from the advantages of each framework,
e.g. the scalability and performance of JAX or the production ecosystem options of TensorFlow.
- Write custom components (e.g. layers, models, metrics) that you can use in low-level workflows in any framework.
- You can take a Keras model and train it in a training loop written from scratch in native TF, JAX, or PyTorch.
- You can take a Keras model and use it as part of a PyTorch-native `Module` or as part of a JAX-native model function.
- Make your ML code future-proof by avoiding framework lock-in.
- As a PyTorch user: get access to power and usability of Keras, at last!
- As a JAX user: get access to a fully-featured, battle-tested, well-documented modeling and training library.
- 在任何框架之上运行高级 Keras 工作流——按需受益于各框架的优势,
例如 JAX 的可扩展性与性能,或 TensorFlow 的生产生态系统选项。
- 编写可在任何框架的低级工作流中使用的自定义组件(例如层、模型、指标)。
- 您可以获取 Keras 模型,并在用原生 TFJAX PyTorch 从头编写的训练循环中训练它。
- 您可以获取 Keras 模型,将其作为 PyTorch 原生 `Module` 的一部分,或作为 JAX 原生模型函数的一部分。
- 通过避免框架锁定,让您的 ML 代码面向未来。
- 作为 PyTorch 用户:终于可以使用 Keras 的强大功能与易用性!
- 作为 JAX 用户:终于可以使用功能齐全、久经考验、文档完善的建模与训练库。
Read more in the [Keras 3 release announcement](https://keras.io/keras_3/).
在 [Keras 3 发布公告](https://keras.io/keras_3/). 中了解更多