如何在requirements.txt/pyproject.toml中为Intel Atom处理器安装定制TensorFlow Wheel？

一、在requirements.txt中实现x86_64无AVX处理器的TensorFlow定制Wheel安装 #

requirements.txt本身不支持直接读取/proc/cpuinfo的复杂硬件检测，但可以通过环境变量+pip条件语法结合辅助脚本实现需求：

1. 修改requirements.txt，添加基于环境变量的条件依赖
   将定制的x86_64无AVX版本TensorFlow相关Wheel加入，用环境变量作为判断开关：

# aarch64架构专属依赖
confluent-kafka @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/v0.1/confluent_kafka-1.9.2-cp38-cp38-linux_aarch64.whl ; platform_machine=='aarch64'
tensorflow @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/v0.1/tensorflow-2.8.4-cp38-cp38-linux_aarch64.whl ; platform_machine=='aarch64'
tensorflow-addons @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/v0.1/tensorflow_addons-0.17.1-cp38-cp38-linux_aarch64.whl ; platform_machine=='aarch64'
tensorflow-text @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/v0.1/tensorflow_text-2.8.2-cp38-cp38-linux_aarch64.whl ; platform_machine=='aarch64'

# x86_64无AVX环境专属依赖（需环境变量NO_AVX=1触发）
tensorflow @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/vx.x/tensorflow-2.8.4-cp38-cp38-linux_x86_64_noavx.whl ; platform_machine=='x86_64' and env NO_AVX='1'
tensorflow-addons @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/vx.x/tensorflow_addons-0.17.1-cp38-cp38-linux_x86_64_noavx.whl ; platform_machine=='x86_64' and env NO_AVX='1'
tensorflow-text @ https://github.com/HandsFreeGadgets/python-wheels/releases/download/vx.x/tensorflow_text-2.8.2-cp38-cp38-linux_x86_64_noavx.whl ; platform_machine=='x86_64' and env NO_AVX='1'

# 通用依赖
rasa==3.4.2
SQLAlchemy==1.4.45
phonetics==1.0.5
de-core-news-md @ https://github.com/explosion/spacy-models/releases/download/de_core_news_md-3.4.0/de_core_news_md-3.4.0-py3-none-any.whl

2. 编写CPU检测前置脚本（check_cpu_avx.py）
   自动检测CPU指令集并设置环境变量，替代直接执行pip install：

import os
import platform
import subprocess

if platform.machine() == 'x86_64':
    try:
        cpu_info = subprocess.check_output("cat /proc/cpuinfo", shell=True).strip()
        # 检测无AVX或手动指定NO_AVX时，开启定制依赖安装
        if b'avx' not in cpu_info or os.environ.get('NO_AVX'):
            os.environ['NO_AVX'] = '1'
        else:
            os.environ['NO_AVX'] = '0'
    except Exception:
        # 检测失败时默认不安装定制版
        os.environ['NO_AVX'] = '0'
else:
    os.environ['NO_AVX'] = '0'

# 触发pip安装
os.system('pip install -r requirements.txt')

3. 使用方式
   运行python check_cpu_avx.py即可完成自动检测与对应依赖安装。

二、具备强安装灵活性的非Legacy Python工具推荐 #

若pyproject.toml的静态条件无法满足需求，推荐以下工具：

• Poetry + 自定义脚本：支持在pyproject.toml中定义前置脚本，将CPU检测逻辑嵌入安装流程，动态调整依赖配置。
• PDM：提供pre_install等钩子脚本，可在安装前执行硬件检测，灵活修改依赖清单。
• Bazel：类似Java生态的Gradle，支持定义复杂构建规则，可实现CPU指令集检测、自定义依赖拉取等高级逻辑，适合复杂项目的构建与依赖管理。

内容的提问来源于stack exchange，提问作者k_o_