"git@developer.sourcefind.cn:renzhc/diffusers_dcu.git" did not exist on "2f24ce1ce3988b2e622bc9a39f29255494bc62a8"
Commit ca5aaf4d authored by zwq330205812's avatar zwq330205812
Browse files

Update README.md

parent 8e1e1917
...@@ -100,6 +100,8 @@ pip install -e . ...@@ -100,6 +100,8 @@ pip install -e .
推理数据集名称:SA-1B Dataset 推理数据集名称:SA-1B Dataset
完整数据集可在[这里](https://ai.facebook.com/datasets/segment-anything-downloads/)进行下载 完整数据集可在[这里](https://ai.facebook.com/datasets/segment-anything-downloads/)进行下载
[快速下载链接](http://113.200.138.88:18080/aidatasets/project-dependency/sa-1b/)
项目中用于试验训练的迷你数据集结构如下 项目中用于试验训练的迷你数据集结构如下
``` ```
── notebooks ── notebooks
...@@ -115,6 +117,7 @@ pip install -e . ...@@ -115,6 +117,7 @@ pip install -e .
### 单机多卡 ### 单机多卡
预训练模型在[这里](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)下载 预训练模型在[这里](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)下载
[快速下载链接](http://113.200.138.88:18080/aimodels/findsource-dependency/sam_vit_h_4b8939/-/raw/master/sam_vit_h_4b8939.pth)
``` ```
git clone https://github.com/luca-medeiros/lightning-sam.git git clone https://github.com/luca-medeiros/lightning-sam.git
cd lightning-sam cd lightning-sam
...@@ -131,6 +134,7 @@ pip install . 过程中会顶掉DCU版本的pytorch、torchvision、triton,需 ...@@ -131,6 +134,7 @@ pip install . 过程中会顶掉DCU版本的pytorch、torchvision、triton,需
python scripts/amg.py --checkpoint <path/to/checkpoint> --model-type <model_type> --input <image_or_folder> --output <path/to/output> python scripts/amg.py --checkpoint <path/to/checkpoint> --model-type <model_type> --input <image_or_folder> --output <path/to/output>
``` ```
注:checkpoint预训练模型在[这里](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)下载 注:checkpoint预训练模型在[这里](https://dl.fbaipublicfiles.com/segment_anything/sam_vit_h_4b8939.pth)下载
[快速下载链接](http://113.200.138.88:18080/aimodels/findsource-dependency/sam_vit_h_4b8939/-/raw/master/sam_vit_h_4b8939.pth)
指令中:<path/to/checkpoint> 代表选择权重的路径 指令中:<path/to/checkpoint> 代表选择权重的路径
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment