Yiming Shi
Yiming Shi
waiting for datasets and training code. :eyes:
use accelerate, accelerator.prepare(model)
@niklasbubeck Hi, my torch version is "torch 2.2.2", maybe you can try the latest torch to see if this works:)
@niklasbubeck Note that you'd better do this once you've load your model ```python # Load pretrained model and tokenizer config = AutoConfig.from_pretrained( model_args.config_name if model_args.config_name else model_args.model_name_or_path, num_labels=num_labels, finetuning_task=data_args.task_name, cache_dir=model_args.cache_dir,...
[Extension Paste Image is not working#25902](https://github.com/orgs/community/discussions/25902#top) @mushanshitiancai I have noticed a similar problem, and a developer mentioned that "Visual Studio Code's API offers access to the local clipboard but only...
> Use Cloudflare Workers to proxy your API and alias the model name. > > ``` > > addEventListener('fetch', event => { > event.respondWith(handleRequest(event.request)) > }) > > const BASE_URL...