Skip to content

Instantly share code, notes, and snippets.

@novaTopFlex
Last active April 19, 2024 18:25
Show Gist options
  • Select an option

  • Save novaTopFlex/25c58bd4d65d3bb5f70f9696229ffeee to your computer and use it in GitHub Desktop.

Select an option

Save novaTopFlex/25c58bd4d65d3bb5f70f9696229ffeee to your computer and use it in GitHub Desktop.
Python Modules in novaTopFlex Projects

Introduction

novaTopFlex believes that modules are absolutely necessary in the vast majority of repositories, including under the future nomenclature system for names of products and projects.

Common Modules in novaTopFlex Repositories (Python)

  • os for system interaction and related purposes/tasks
  • sys for system information and monitoring
  • tkinter for basic graphical computing
  • wx for wxPython (advanced graphical interfacing)
  • tensorflow for TensorFlow artificial intelligence

Recommended Installation Procedures for wxPython and TensorFlow

  • wxPython: pip install wxPython
  • TensorFlow: pip install tensorflow for systems with NVIDIA Graphics or pip install tensorflow-cpu for systems with integrated or non-NVIDIA dedicated graphics chips

TensorFlow: GPU or CPU?

  • For systems with NVIDIA Graphics, novaTopFlex recommends the GPU-enabled package for TensorFlow.
  • For laptops with dual-graphics (Integrated plus NVIDIA), novaTopFlex generally recommends the GPU-enabled package for TensorFlow; however, when the laptop is in integrated graphics mode, some features may be broken. When this occurs, novaTopFlex recommends switching, at least temporarily, to the TensorFlow CPU version.
  • For systems with dedicated graphics (not NVIDIA), novaTopFlex recommends the CPU-only package for TensorFlow; the CUDA operations of the GPU package are designed specifically with NVIDIA use cases in mind.
  • For systems with integrated graphics only, noveTopFlex recommends the CPU-only package for TensorFlow.
  • For QEMU virtual machines with KVM acceleration enabled, novaTopFlex recommends the CPU-only package for TensorFlow, regardless of the technical specifications of the host system. Although one's system may contain NVIDIA Graphics, the virtual machine would not recognize the graphics card to the same extent, and when the VGA device is incorporated, the megabytes of VGA memory are to be inherited from the system (CPU), not from the GPU.

Tkinter or wxPython?

While both the tkinter (formerly Tkinter in Python 2) and wx modules provide GUI support to the Python programming language, the specific purposes being served are somewhat different. Generally, simpler applications are more suitable for Tkinter, and more complex interfaces are more suitable for wxPython. The wxPython standard library contains more widgets than in Tkinter; however, that often comes with a caveat: Multiple widget types may seem nearly identical with certain variations becoming different widgets while these variations remain the same widget under Tkinter.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment