ai-voice-cloning/train.sh

#!/bin/bash
source ./venv/bin/activate

GPUS=$1
CONFIG=$2
PORT=1234

if (( $GPUS > 1 )); then
	torchrun --nproc_per_node=$GPUS --master_port=$PORT ./src/train.py -opt "$CONFIG" --launcher=pytorch
else
	python3 ./src/train.py -opt "$CONFIG"
fi
deactivate
crunchbangs 2023-02-18 02:46:44 +00:00			`#!/bin/bash`
a bit of UI cleanup, import multiple audio files at once, actually shows progress when importing voices, hides audio metadata / latents if no generated settings are detected, preparing datasets shows its progress, saving a training YAML shows a message when done, training now works within the web UI, training output shows to web UI, provided notebook is cleaned up and uses a venv, etc. 2023-02-18 02:07:22 +00:00			`source ./venv/bin/activate`
experimental multi-gpu training (Linux only, because I can't into batch files) 2023-03-03 04:37:18 +00:00
			`GPUS=$1`
			`CONFIG=$2`
			`PORT=1234`

			`if (( $GPUS > 1 )); then`
use torchrun instead for multigpu 2023-03-04 20:53:00 +00:00			`torchrun --nproc_per_node=$GPUS --master_port=$PORT ./src/train.py -opt "$CONFIG" --launcher=pytorch`
experimental multi-gpu training (Linux only, because I can't into batch files) 2023-03-03 04:37:18 +00:00			`else`
			`python3 ./src/train.py -opt "$CONFIG"`
			`fi`
a bit of UI cleanup, import multiple audio files at once, actually shows progress when importing voices, hides audio metadata / latents if no generated settings are detected, preparing datasets shows its progress, saving a training YAML shows a message when done, training now works within the web UI, training output shows to web UI, provided notebook is cleaned up and uses a venv, etc. 2023-02-18 02:07:22 +00:00			`deactivate`