你好,我是坚持分享干货的 EarlGrey,翻译出版过《Python编程无师自通》、《Python并行计算手册》等技术书籍。
如果我的分享对你有帮助,请关注我,一起向上进击。
PyPI(Python Package Index)是一个用于发布和管理 Python 包的平台,它提供了方便的安装,更新,和搜索功能,让 Python 开发者可以轻松地获取和使用各种各样的 Python 库和框架。PyPI 上有近 50 万个 Python 包,涵盖了各个领域和应用场景,是 Python 生态系统的重要组成部分。
但是,你有没有想过,PyPI 上的 Python 包有多大?所有包加起来有多大?
据PyPI官网数据,目前所有发布在PyPI上的包(包括所有版本),合计大小有 19 TB之多,最大的包是 TensorFlow,历史所有版本的包加在一起有 313.1 GB,紧随其后大小超过200 GB 的包有 tf-nightly-intel、lalsuite、tf-nightly-cpu-aws、tensorflow-gpu、catboost-dev、tensorflow-io-nightly。
可以看到,排名靠前的包里,绝大部分都是和 tensorflow 有关的。
下面是包大小前100的包名称和对应大小,供大家参考。
Project name | Sum of release files (bytes) |
---|---|
All of PyPI | 19.0 TB |
tensorflow | 313.1 GB |
tf-nightly-intel | 306.2 GB |
lalsuite | 262.6 GB |
tf-nightly-cpu-aws | 246.3 GB |
tensorflow-gpu | 237.9 GB |
catboost-dev | 237.0 GB |
tensorflow-io-nightly | 205.4 GB |
tf-nightly | 196.3 GB |
paddlepaddle-gpu | 184.1 GB |
OpenVisus | 166.8 GB |
frida | 160.5 GB |
tf-nightly-macos | 139.9 GB |
tensorflow-cpu | 137.6 GB |
torch | 131.3 GB |
tf-nightly-cpu | 97.0 GB |
tensorflow-rocm | 74.8 GB |
cupy-cuda92 | 71.2 GB |
cupy-cuda100 | 70.2 GB |
grpcio | 67.7 GB |
ddtrace | 67.3 GB |
cupy-cuda90 | 66.0 GB |
pyAgrum-nightly | 62.9 GB |
cupy-cuda91 | 55.5 GB |
cupy-cuda101 | 55.0 GB |
grpcio-tools | 54.5 GB |
catboost | 54.2 GB |
ray | 53.0 GB |
ovito | 49.4 GB |
mindspore-dev | 49.2 GB |
opencv-contrib-python | 46.9 GB |
scipy | 46.5 GB |
opencv-contrib-python-headless | 42.7 GB |
opencv-python | 42.7 GB |
pantsbuild.pants | 38.7 GB |
deepspeech-gpu | 36.6 GB |
cupy-cuda80 | 36.3 GB |
cupy-cuda102 | 35.5 GB |
cmake | 35.2 GB |
sickrage | 34.2 GB |
mindspore | 33.4 GB |
numpy | 32.9 GB |
Panda3D | 32.2 GB |
pyqt5-tools | 32.2 GB |
opencv-python-headless | 32.0 GB |
codeforlife-portal | 31.8 GB |
pulumi-azure-native | 31.7 GB |
udata | 31.5 GB |
pulsar-client-sn | 31.2 GB |
tensorflow-io-2.0-preview | 29.3 GB |
paddlepaddle | 28.7 GB |
tensorflow-intel | 27.7 GB |
tf-gpu | 27.0 GB |
azureml-dataprep-rslex | 26.3 GB |
pybullet | 25.4 GB |
cupy-cuda111 | 24.0 GB |
CodeIntel | 23.8 GB |
botocore | 23.7 GB |
pygame | 23.6 GB |
taichi | 23.3 GB |
casadi | 23.0 GB |
awscrt | 22.0 GB |
matplotlib | 21.6 GB |
jaxlib | 21.4 GB |
pyarrow | 21.4 GB |
homeassistant | 21.4 GB |
aim | 21.3 GB |
tendenci | 21.3 GB |
zeroconf | 21.1 GB |
intel-tensorflow | 21.1 GB |
rasterio | 21.0 GB |
suitesparse-graphblas | 20.9 GB |
apache-flink | 20.9 GB |
databricks-connect | 20.7 GB |
itk-filtering | 20.4 GB |
PySide2 | 20.2 GB |
chdb | 20.0 GB |
cupy-cuda110 | 20.0 GB |
xpress | 20.0 GB |
monocdk | 19.9 GB |
aws-cdk-lib | 19.4 GB |
itk-core | 19.3 GB |
polars | 19.2 GB |
spacy | 19.1 GB |
tensorflow-cpu-aws | 18.9 GB |
ray-cpp | 18.7 GB |
tensorflow-macos | 18.4 GB |
xformers | 18.4 GB |
apache-beam | 18.3 GB |
home-assistant-frontend | 18.0 GB |
onnxruntime-gpu | 17.9 GB |
cdktf-cdktf-provider-aws | 17.7 GB |
ctranslate2 | 17.6 GB |
open3d | 17.5 GB |
intel-extension-for-tensorflow-lib | 17.5 GB |
allennlp-pvt-nightly | 17.5 GB |
ruff | 17.3 GB |
pandas | 17.1 GB |
kolibri | 17.1 GB |
semgrep | 17.0 GB |
tiledb | 16.8 GB |
- EOF -
文章已经看到这了,别忘了在右下角点个“赞”和“在看”鼓励哦~
告诉你更多细节干货
欢迎围观我的朋友圈
每天更新所想所悟