Server/fish-speech @ 2733ed600ff90bfb38d3fcf3eee2bf173c02f449

Nessuna descrizione

Lengyue 2733ed600f Accelerate training with fsdp		2 anni fa
preparing_data	9adf687fb1 Update base training code	2 anni fa
speech_lm	2733ed600f Accelerate training with fsdp	2 anni fa
.dockerignore	70797ef7d5 Add dockerignore	2 anni fa
.gitignore	7650b2ac43 Update much robust dataloader	2 anni fa
.pre-commit-config.yaml	e4ce816175 Update deps	2 anni fa
.project-root	aa0a6be778 init training code	2 anni fa
README.md	e4ce816175 Update deps	2 anni fa
dockerfile	7650b2ac43 Update much robust dataloader	2 anni fa
ds_config.json	db0aa1a99a Init preprocessing tools	2 anni fa
requirements.txt	2733ed600f Accelerate training with fsdp	2 anni fa
setup.py	7650b2ac43 Update much robust dataloader	2 anni fa
train.sh	a0a5c3e95c Update training script	2 anni fa

		
				README.md
			
				Speech LLM

Setup

# Basic environment setup
conda create -n speech-llm python=3.10
conda activate speech-llm
conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

# Install requirements
pip3 install -r requirements.txt

# Install flash-attn
MAX_JOBS=4 pip install flash-attn --no-build-isolation