問題描述
我正在嘗試測試 我的 Ultimate-utils 庫的安裝,我得到了這個奇怪的錯誤:
I was trying to test the installation of my ultimate-utils library and I get this bizzare error:
(meta_learning) brandomiranda~/ultimate-utils ? pip install -e .
Obtaining file:///Users/brandomiranda/ultimate-utils
Preparing metadata (setup.py) ... done
Collecting dill
Using cached dill-0.3.4-py2.py3-none-any.whl (86 kB)
Collecting networkx>=2.5
Using cached networkx-2.6.3-py3-none-any.whl (1.9 MB)
Requirement already satisfied: scipy in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (1.7.3)
Collecting scikit-learn
Using cached scikit_learn-1.0.1-cp38-cp38-macosx_10_13_x86_64.whl (7.9 MB)
Collecting lark-parser
Using cached lark_parser-0.12.0-py2.py3-none-any.whl (103 kB)
Collecting torchtext==0.10.1
Using cached torchtext-0.10.1-cp38-cp38-macosx_10_9_x86_64.whl (1.6 MB)
Collecting tensorboard
Using cached tensorboard-2.7.0-py3-none-any.whl (5.8 MB)
Collecting pandas
Using cached pandas-1.3.4-cp38-cp38-macosx_10_9_x86_64.whl (11.4 MB)
Collecting progressbar2
Using cached progressbar2-3.55.0-py2.py3-none-any.whl (26 kB)
Collecting transformers
Using cached transformers-4.13.0-py3-none-any.whl (3.3 MB)
Requirement already satisfied: requests in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (2.26.0)
Collecting aiohttp
Using cached aiohttp-3.8.1-cp38-cp38-macosx_10_9_x86_64.whl (574 kB)
Requirement already satisfied: numpy in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (1.21.4)
Collecting plotly
Using cached plotly-5.4.0-py2.py3-none-any.whl (25.3 MB)
Collecting wandb
Using cached wandb-0.12.7-py2.py3-none-any.whl (1.7 MB)
Collecting matplotlib
Using cached matplotlib-3.5.0-cp38-cp38-macosx_10_9_x86_64.whl (7.3 MB)
Requirement already satisfied: torch<1.10.0,>=1.4.0 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (1.9.1)
Requirement already satisfied: torchvision<0.11.0,>=0.5.0 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (0.10.1)
Requirement already satisfied: torchaudio<0.10.0,>=0.9.1 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (0.9.1)
ERROR: Package 'ultimate-utils' requires a different Python: 3.8.1 not in '>=3.9.0'
但是,如果您檢查依賴項似乎已正確安裝 - 所以我很困惑這條消息的含義.
however, if you check the dependencies seem to have installed correctly - so I am puzzled what this message means.
此外,如果您更仔細地閱讀該消息,您還會看到它出于某種原因引用了 3.8(python 3.8?不確定),請參閱:
In addition if you read the message more carefully you also see that it references 3.8 for some reason (python 3.8? not sure), see:
Requirement already satisfied: torch<1.10.0,>=1.4.0 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (1.9.1)
Requirement already satisfied: torchvision<0.11.0,>=0.5.0 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (0.10.1)
Requirement already satisfied: torchaudio<0.10.0,>=0.9.1 in /Library/Frameworks/Python.framework/Versions/3.8/lib/python3.8/site-packages (from ultimate-utils==0.5.3) (0.9.1)
ERROR: Package 'ultimate-utils' requires a different Python: 3.8.1 not in '>=3.9.0'
但如果我要求 pythons 版本,它是我期望的 3.9:
but if I ask for pythons version its the one I expected 3.9:
(meta_learning) brandomiranda~/ultimate-utils ? python
Python 3.9.7 (default, Sep 16 2021, 08:50:36)
[Clang 10.0.0 ] :: Anaconda, Inc. on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>>
所以我很困惑到底發生了什么.我想知道pycharm是否與此有關.但是為什么說它沒有安裝我的 uutils 庫呢?
So I am very puzzled what is going on. I wonder if pycharm has anything to do with this. But why is it saying it did not install my uutils library?
檢查點子列表確認它不存在:
and checking the pip list confirms its not there:
(meta_learning) brandomiranda~/ultimate-utils ? pip list
Package Version
---------------------- ---------
argcomplete 1.12.3
astroid 2.3.3
boto 2.49.0
certifi 2021.10.8
cffi 1.15.0
charset-normalizer 2.0.9
cloudpickle 2.0.0
crcmod 1.7
cryptography 36.0.0
fasteners 0.16.3
gcs-oauth2-boto-plugin 3.0
google-apitools 0.5.32
google-reauth 0.1.1
gsutil 5.5
gym 0.21.0
httplib2 0.20.2
idna 3.3
isort 4.3.21
lazy-object-proxy 1.4.3
learn2learn 0.1.6
mccabe 0.6.1
monotonic 1.6
nodejs 0.1.1
numpy 1.21.4
oauth2client 4.1.3
optional-django 0.1.0
Pillow 8.4.0
pip 21.3.1
pyasn1 0.4.8
pyasn1-modules 0.2.8
pycparser 2.21
pylint 2.4.4
pyOpenSSL 21.0.0
pyparsing 3.0.6
pyu2f 0.1.5
qpth 0.0.15
requests 2.26.0
retry-decorator 1.1.1
rsa 4.7.2
scipy 1.7.3
setuptools 41.2.0
six 1.14.0
torch 1.9.1
torchaudio 0.9.1
torchvision 0.10.1
tqdm 4.62.3
typing_extensions 4.0.1
urllib3 1.26.7
wrapt 1.11.2
從 uutils 運行導入確認它沒有安裝:
running an import from uutils confirms it was not installed:
(meta_learning) brandomiranda~/ultimate-utils ? python -c "import uutils; uutils.hello()"
python -c "import uutils; uutils.torch_uu.hello()"
Traceback (most recent call last):
File "<string>", line 1, in <module>
ModuleNotFoundError: No module named 'uutils'
Traceback (most recent call last):
File "<string>", line 1, in <module>
ModuleNotFoundError: No module named 'uutils'
奇怪的是,這在 python 3.9 之前已經有效.所以有些東西改變了,我不知道是什么.
what is bizzare is that this has worked before in python 3.9. So something changed and I can't figure out what.
這一定是python路徑問題,因為現在它在刪除python 3.8(我從未安裝過......)后打開了2.7.我認為 conda init 沒有正確啟動我的 shell.在哪里:
it must be a python path issue since now it's opening 2.7 after deleting python 3.8 (which I never installed...). I think conda init is not initiating my shell correctly. Where is:
# >>> conda initialize >>>
# !! Contents within this block are managed by 'conda init' !!
__conda_setup="$('/Users/brandomiranda/miniconda3/bin/conda' 'shell.zsh' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/brandomiranda/miniconda3/etc/profile.d/conda.sh" ]; then
. "/Users/brandomiranda/miniconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/brandomiranda/miniconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
假設進入.zshrc
文件?
參考/相關鏈接:
- 無法安裝新的適用于 Python 的軟件包(Python 3.9.0、Windows 10)
- 為什么 Pip 聲稱 Python 的版本不在給定范圍內?
- 如何是否從 setup.py install_requires 列表中安裝 PyTorch 和相關工具?
- https://github.com/brando90/ultimate-utils
- Python OSX $ 其中 Python 提供/庫/Frameworks/Python.framework/Versions/2.7/bin/python
- 為什么 conda init 更新我的 .bash_profile 不正確?
- 如何從 macOS 完全卸載 Anaconda
- https://superuser.com/questions/121870/how-do-i-reset-the-path-variable-on-mac-os-x
- https://github.com/brando90/.dotfiles/blob/master/README.md
推薦答案
一旦我開始使用python 3.9(而不是3.10,可能使用最新版本的python 是一個壞主意,所以我學會了只落后大多數軟件版本以避免奇怪的錯誤)然后我確保我的 mac 的新默認 shell zsh
有一個全新的原始路徑:
Once I started using python 3.9 (and not 3.10, probably using the most recent version of python is a bad idea, so I learned to be just 1 behind most version of software to avoid weird bugs) and then I made sure my mac's new default shell zsh
has a brand new virgin path:
PATH=/usr/bin:/bin:/usr/sbin:/sbin
export PATH
在它的頂部.
# In my 2013 mac this seems to be needed
PATH=/usr/bin:/bin:/usr/sbin:/sbin
export PATH
# let zplug manage itself
zplug "zplug/zplug"
## Plugins for zsh
# new `jump`
zplug "plugins/wd", from:oh-my-zsh
## prompt
## to test do: print -P '%B%F{red}co%F{green}lo%F{blue}rs%f%b'
# ref: http://zsh.sourceforge.net/Doc/Release/Prompt-Expansion.html#Prompt-Expansion
# ref: reference: https://wiki.archlinux.org/index.php/Zsh#Colors
# ref: https://scriptingosx.com/2019/07/moving-to-zsh-06-customizing-the-zsh-prompt/
# %F{color} sets colour until next color of %f, %B infront of that changes tone
PROMPT="%B%F{magenta}%n%B%F{red}%~%B%F{green} %F{cyan}?%f "
## jump TODO: switch to zsh's wd
#conda activate metalearning
# conda activate synthesis
conda activate meta_learning
## Aliases
alias grep='grep --color=auto'
alias intel='ssh -i ~/.ssh/intel_id_rsa miranda9@ssh-iam.intel-research.net'
alias hal='ssh miranda9@hal.ncsa.illinois.edu'
alias vision='ssh miranda9@vision-submit.cs.illinois.edu'
alias nano='ssh miranda9@nano.ncsa.illinois.edu'
alias dgx='ssh miranda9@hal-dgx.ncsa.illinois.edu'
alias sshcc='ssh miranda9@cc-login.campuscluster.illinois.edu'
#alias tb='tensorboard --logdir'
alias tbb='sh /Users/brando/ultimate-utils/run_tb.sh'
export WANDB_API_KEY=MY SCRETE MUAHAHAHA
# >>> conda initialize >>>
# !! Contents within this block are managed by 'conda init' !!
__conda_setup="$('/Users/brandomiranda/opt/anaconda3/bin/conda' 'shell.zsh' 'hook' 2> /dev/null)"
if [ $? -eq 0 ]; then
eval "$__conda_setup"
else
if [ -f "/Users/brandomiranda/opt/anaconda3/etc/profile.d/conda.sh" ]; then
. "/Users/brandomiranda/opt/anaconda3/etc/profile.d/conda.sh"
else
export PATH="/Users/brandomiranda/opt/anaconda3/bin:$PATH"
fi
fi
unset __conda_setup
# <<< conda initialize <<<
我還重新安裝了 python,刪除了 python 框架 3.8 并重新安裝了 conda 作為圖形界面,但我不認為這些是讓事情起作用的東西.我相當有信心這是正確的.
I also reinstalled python, removed python framework 3.8 and reinstalled conda as a graphical interface but I don't think these were the things that made things work. I am fairly confident that is right.
這篇關于為什么 python 使用 3.8.1 和 3.9,然后無法安裝包(錯誤:包 pkg 需要不同的 Python:3.8.1 不在 '>=3.9.0' 中)?的文章就介紹到這了,希望我們推薦的答案對大家有所幫助,也希望大家多多支持html5模板網!