-
The official mentioned that blip uses a resolution of 224, which may not be good for image detail understanding, can finetune training modify the image size? If not, and need to train from zero, then …
-
![f7cd7b90cbea771093ef0a2dea40a37](https://github.com/THUDM/VisualGLM-6B/assets/36267615/ea42bb6a-62be-475c-b6e9-d13cabb822ee)
![70bef0aa692a65f3959727504d7538b](https://github.com/THUDM/VisualGLM-6B…
-
如题
-
在运行cli_demo.py进行部署时报错如下:
[2023-07-24 06:19:37,012] [INFO] [real_accelerator.py:133:get_accelerator] Setting ds_accelerator to cuda (auto detect)
/remote-home/cwli/anaconda3/envs/minigpt4/lib/python3…
-
![Snipaste_2023-10-19_15-41-33](https://github.com/THUDM/VisualGLM-6B/assets/147784112/ad95e8f9-7951-4634-829f-87001c81187a)
这是cuda版本算力问题吗
-
0.2版本能使用多模多模型吗?比如:VisualGLM-6B
-
尝试安装时候出现以下错误
```
PS D:\deeplearning\VisualGLM-6B> pip install SwissArmyTransformer>=0.4.4
Collecting SwissArmyTransformer>=0.4.4
Obtaining dependency information for SwissArmyTransformer>=0.4.4 …
-
-
环境:nvidia a10 24g显存,docker:nvidia/cuda:11.8.0-cudnn8-devel-ubuntu20.04,cpu:Intel® Xeon® Silver 4314×2,mem:256G
日志如下:
NCCL_DEBUG=info NCCL_IB_DISABLE=0 NCCL_NET_GDR_LEVEL=2 deepspeed --master_port 16…
-
如题,使用的虚拟环境,已经安装所需依赖。在 HuggingFace 上下载的模型,放置到本地文件夹 /data/models/THUDM/visualglm-6b 中。修改 cli_demo.py:
```
def main():
...
# load model
model, model_args = VisualGLMModel.from_pretrained…