在用torch.distributed.init_process_group(backend='nccl', init_method='env://', world_size=world_size, rank=rank)
时,出现
1、ValueError: Error initializing torch.distributed using env:// rendezvous: environment variable MASTER_ADDR expected, but not set
解决
加入
os.environ['MASTER_ADDR'] = 'localhost'
2、ValueError: Error initializing torch.distributed using env:// rendezvous: environment variable MASTER_PORT expected, but not set
解决
加入
os.environ['MASTER_PORT'] = '12345'
- 分布式 initializing distributed swin-trans ValueError分布式initializing distributed swin-trans 分布式distributed zookeeper chatgpt swin-trans valueerror non-boolean valueerror containing boolean llamatokenizer valueerror tokenizer currently valueerror wordcloud supported truetype draw draw_rectangle valueerror rectangle valueerror python问题 quot logging_dir valueerror unbuffered