Scripts for fine-tuning Meta Llama with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom datasets for applications such as summarization and Q&A. Supporting a number of candid inference solutions such as HF TGI, VLLM for local or cloud deployment. Demo apps to showcase Meta Llama for WhatsApp & Messenger.
What does this PR do?
Fix NotebookLlama demo
warnings
from requirements, since we can directlyimport warnings
in python 3wavfile
,AudioSegment
andio
to 'Step-4-TTS-Workflow.ipynb'pydub
andparler-tts
required from 'Step-4-TTS-Workflow.ipynb'transformers
version to 4.43.0, to be compatible withparler-tts
's dependencyFeature/Issue validation/testing
not applicable