mongodb: mongodump -out backup/ -> mongorestore backup/.restore MySQL/Postgres/MongoDB databases.

create virtual environments for projects mkvirtualenv and run pip install -r requirements.txt.Mkvirtualenv3 project_name Django completionĪdd autocompletion in terminal when we type manage.py or django-admin.py and press button two times. # you can also use my shortcut from ~/.profile (see link to file above) Mkvirtualenv -python= $(which python3 ) project_name Then run next command to make this directory writable: It happens that permissions on some files and directories broken after upgrade to newer version of macOS. Repair disk permissions with Disk Utility. Find it by the name fix_virtualenv Bad file permissions Use it like this fix_virtualenv and it will auto-fix your python version by replacing a broken links to an actual version of python.įind the snippet in my gist. I've created an alias that is easy to use every time when you upgrade the python version via homebrew, and your virtusl env becomes not working. How to fix a broken virtual env after python version upgrade? Known problems Virtual env stopped work when I upgraded python via homebrew How to clean unused homebrew dependencies.Virtual env stopped work when I upgraded python via homebrew.When I upgrade to a next major version of macOS it's almost always some problems appear - some tools stop working, eapecially if you use your system for software development. Sudo port install p圓6-graphviz … is it right?īut I see that they install dependencies like python 3.On this page I describe how to setup python environment on macOS Catalina (10.13). Here are my experience that I share according to your instructions:ġ) after installing MacPorts …you have to close and reopen terminal in order to get the right answer of updated to the command : ‘sudo port selfupdate’Ģ) I though Anaconda or Miniconda take better care of modules dependencies installation, but the way you propose through MacPorts you have to take car of it…is it right?ģ) to me installing Keras and tensorflow is the base for my software, but Keras said it required that the only current python compatibility is 2.7 or 3.6 version …so installing 3.5 python must be corrected from this tutorial to 3.6Ĥ) in keras page installation to other software are recommended for installation:Ĥ.1) h5py, that I understand it is already included in pyhton 3.6…nothing additional to do itĤ.2) pydot (for visual python utilities)…but because they recommended install by pip … I decided to do the same command installation structure as you do for scikit-leanr with pip :1Ĥ.3) regarding graphviz (also a visual utility) I follow the Macport recommedation installation: Motivation: Due to recent mac os upgrade from Mojave to Catalina 10.15 they uninstall anaconda from root directories and because I used the old 2.7 anaconda I decided to reinstall all my python+tensorflow+keras environment according to this tutorials.

Python mac os