What This is an article that summarizes what I noticed and researched when learning machine learning using Chainer. This time, I will study scikit-learn.
It is written based on my understanding, so it may be incorrect. I will correct any mistakes, please forgive me.
Content
GPU An arithmetic unit specialized in image processing. I have no idea what is different from the CPU in principle and is advantageous for image processing. For example, the following site briefly explains the difference between CPU and GPU. https://www.datadock.co.jp/column/GPU/2018/05/88.html However, in order to understand the essentials, it seems necessary to first suppress the processing that the CPU is good at and the processing that the GPU is good at. To tell the truth, it is currently the area where parallel computing can speed up the process of performing multiple similar processes at the same time. I only found that the GPU can process numerical calculations in parallel and at high speed. By the way, GPU maker NDIVIA has been secretly attracting attention recently, and it seems that the flow of AI is big. ** Cupy requires a GPU, but Colab can also use a GPU, which is amazing! ** **
Cupy Even if I read the contents, honestly, I can't summarize it so much. .. .. By writing in Cupy, numerical calculation can be performed on the GPU, and high-speed operation can be expected. Numerical calculations are written using Cupy instead of Numpy as much as possible, and code that operates at high speed is applied. If you have trouble writing, it seems okay to check it as appropriate
Comment It's a time when the temperature drops suddenly and it's easy to get sick ... I fell asleep on Saturdays and Sundays due to severe headaches and drowsiness orz
Recommended Posts