đ´2. Optional: TensorBoard (RVC)
Just in case...
A simple program that automates the search for the optimal number of epochs for your RVC model.
I. Python Installation
Go to the Python download page.
Click on the "
Download Python
" button.Once the download is complete, run the installation file and follow the instructions to install Python.
During the installation process, make sure to check the boxes next to "Add Python to PATH" and "Install IDLE."
II. Installation of ROD
Go to the ROD download page.
Under Assets, click on the first file (ROD-v...).
Unzip the downloaded file to the location of your choice.
III. Local Training:
If you plan to train locally, create a directory named "ROD
" within your "Mangio-RVC
." Place the extracted files inside the created "ROD
" directory.
IV. Install Prerequisites:
Run the "install-requirements.bat
" script to install the necessary dependencies.
V. Create the Configuration File:
Use the "config-editor.bat
" script to create and edit a configuration file according to your needs.
VI. Customize Your Configuration:
The model name serves as an identifier for monitoring the specified model. Iterations determine the number of epochs that need to follow the lowest point on the graph. For example, if you choose 100 iterations, and there are only 20 epochs after the lowest point, the refresh will occur after the specified number of seconds. If you already have a trained model, set iterations to 1. The refresh rate indicates how often the program will update in seconds, if no value is provided.
VII. Launch Tensorboard:
Start Tensorboard by running the "launch-tensorboard.bat
" script.
VIII. Run rod.bat:
Execute the "rod.bat
" script and obtain the result of the sufficient number of epochs.
IX. How to Use TensorBoard:
Click on the "Scalars
" tab and look for "g/total
" at the top.
Last updated