# Mainline Colab

Last update: Oct 23, 2024

Mainline colab is a port of mainline RVC to Google Colab, for exclusively training.
It's free, includes all the necessary tools for a quality model, the TensorBoard.

‎

# Pros & Cons

The pros & cons are subjective to your necessities.

✔️ PROS:

❌ CONS

Has TensorBoard.

Inconvenient.
Takes some time to set up.
You can't leave training unsupervised.
Doesn't have Mangio-crepe.
For free users:
- It's slower compared to local RVC.
- Can't train long datasets without pausing the process.

# ‎

# How to Setup

# ‎

# 1. Running cells.

Start by accessing the colab here.
Then run the first two cells to install all the requirements.

#

# 2. Installing Pretrains.

If you wish to install a custom pretrain go to the 'Download Custom Pretrains' cell and go into the dropdown menu and find the pretrain you want.
If the pretrain you want isn't there go to the top left and click '+ Code'.
Then in the new cell type in !wget LINK TO PRETRAIN

# ‎

# 3. Ngrok.

Scroll down to the fifth cell and you should see a section where you put your ngrok token. If you don't have a ngrok acount sign up here.
‎
2. Once you have an acount you can authenticate your ngrok tunnel agent here: https://dashboard.ngrok.com/get-started/your-authtoken
‎
put the Ngrok token in like so:

‎

Warning

There is a monthly limit rate with Ngrok so don't be supprised if training is suddenly interrupted.

#

# 3. The Imjoy Elfinder WebGUI.

The interface should look like this with your D and G files being located here. Here you can manage your files. Whenever you want to download files from the Imjoy GUI just double click. Downloading files uses up the Ngrok bandwith data.

# ‎

# Starting RVC

# ‎

# 1. Click the RVC link.

It should take you to the GUI where you can then go to the top and click the 'Train' tab.

# 2. Setup.

Run through the normal RVC setup with setting your model name, sample rate and such. If you are unable to see the 32k sample rate click on V1 then swap back to V2.

Only use V2 Don't use V1 or you will get an error.

In google drive create a folder named training and inside it make another folder nameddataset then drag and drop your dataset in it. Then continue with normal RVC setup and training

Make sure you use WAV or Flac files inside. Do not zip the folder.

# 3. Syncing Graphs.

For syncing graphs you need to train however many epochs you have set you save frequency then go into the file manager and find your model which should be in training/weights, in its name it should have a step count and epoch count like this: model_name_e(number)_s(number).

E means epochs and S means steps.

Once you know how many steps the model trained for stop training the model by stopping the cell.

Then start the same cell and open then navigate to assest/weights in google colab and delete all of your previous models.
Then navigate to the 32k.json file which is located in the V2 folder of /configs and edit log_interval to the amount of steps your model took, save it.

Now go to your /logs/ folder and do the same thing. Modify the log interval of the config.json with your step count.

Now delete Eval folder, tf-events file, G & D_23333333 files and train log file in /logs/ folder.

# 4. Resuming Training.

Do not process and feature extract again because those files are already in the /logs/ folder. Use the same model name, sample rate, batch size, pretrain, and save frequency to train the model again in the RVC GUI.

# ‎ ‎

# `You have reached the end.`

Report Issues