osu! Beatmap Generator Installation Tutorial (Using Machine Learning) смотреть онлайн

This video focuses on how to install and get the generator working just in case you don't know what to do when you first came across the reddit post. This generator uses Tensorflow / Deep Learning to generate and create an osu! beatmap.

Important Details: Make sure in the .ipynb files, when you are editing the directory, make sure you remember to an in another "\" so it'll look something like this "C:\\ffmpeg\\bin\\ffmpeg.exe" instead of "C:\ffmpeg\bin\ffmpeg.exe". Same thing for locating the directory for the empty .osu file for your diff. (DO NOT USE THE DOUBLE \ IN THE MAPLIST.TXT! ONLY IN THE .ipynb FILES)

Original Reddit Post: https://www.reddit.com/r/osugame/comments/98j74a/an_automatic_beatmap_generator_using_tensorflow/

Download Links:
GitHub - https://github.com/kotritrona/osumapper
Notepad++ - https://notepad-plus-plus.org/download/
Python 3.6.6 - https://python.org/downloads/release/python-366/
Node.js - https://nodejs.org/en/download/
FFmpeg - http://ffmpeg.org/download.html#build-windows (NEW LINK, OLD DOWNLOAD DEAD)

Commands for cmd:
npm i polynomial
pip install soundfile
python -m pip install jupyter
pip install pandas
pip install numpy
pip install matplotlib
pip install h5py

IMPORTANT FOR INSTALLING TENSORFLOW:
FOR TENSORFLOW INSTALLATION, IT DEPENDS IF YOU HAVE A COMPATIBLE NVIDIA GRAPHIC CARD. IF YOU DO, TYPE IN "pip3 install --upgrade tensorflow-gpu"

For more information for installing Tensorflow for an NVIDIA graphic card, check this link out - https://www.tensorflow.org/install/install_windows

IF YOU DONT HAVE AN NVIDIA GRAPHIC CARD/HAVE AN AMD GPU OR YOU JUST WANT TO USE YOUR CPU, THEN TYPE IN "pip3 install --upgrade tensorflow" IN CMD.

If you are getting this error while installing Tensorflow, "Could not install packages due to an EnvironmentError: [WinError 5] Access is denied:", then type in (in you have an AMD gpu or if you just want to use your cpu) "pip3 install tensorflow" or otherwise, if you have an Nvidia gpu, then type in "pip3 install tensorflow-gpu".

To launch .ipynb, type in "jupyter notebook" in CMD and wait for your browser to launch and locate the osumapper-master folder that contains your .ipynb files.

Here's 2 maps that I've created using GAN (Deep Learning):
https://osu.ppy.sh/beatmapsets/835901/
https://osu.ppy.sh/beatmapsets/818053/