search

THUDM / VisualGLM-6B

Chinese and English multimodal conversational language model | 多模态中英双语对话语言模型
Apache License 2.0
4.07k stars 414 forks source link

要设置WORLD_SIZE and LOCAL_RANK #303

Open doublehelix626 opened 10 months ago

doublehelix626 commented 10 months ago

(visualglm) D:\visuaglm\VisualGLM-6B>python api.py [2023-10-29 14:47:17,801] [INFO] DeepSpeed/CUDA is not installed, fallback to Pytorch checkpointing. [2023-10-29 14:47:17,974] [WARNING] DeepSpeed Not Installed, you cannot import training_main from sat now. [2023-10-29 14:47:18,524] [INFO] building VisualGLMModel model ... [2023-10-29 14:47:18,539] [INFO] [RANK 0] > initializing model parallel with size 1 [2023-10-29 14:47:18,539] [INFO] [RANK 0] You are using model-only mode. For torch.distributed users or loading model parallel models, set environment variables RANK, WORLD_SIZE and LOCAL_RANK.

doublehelix626 commented 10 months ago

请问怎么解决这个问题？

JaysonCai236 commented 6 months ago

请问解决该问题了吗

doublehelix626 commented 5 months ago

你好，这个问题解决了，这个并不是警告或者是报错，只是某种参数信息的显示，之后就会开始加载权重参数并开始模型推理，如果没有开始加载权重参数那说明显存可能不够，没法进行下一步

在 2024-02-26 19:59:13，"Jayson" @.***> 写道：

请问解决该问题了吗

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>