Skip to content

Commit dce5030

Browse files
committed
Update README.md to enhance CUDA installation instructions and add details for CUDA 13 support
1 parent ede00fe commit dce5030

File tree

1 file changed

+39
-7
lines changed

1 file changed

+39
-7
lines changed

README.md

Lines changed: 39 additions & 7 deletions
Original file line numberDiff line numberDiff line change
@@ -64,44 +64,76 @@ diffct/
6464

6565
### Installation
6666

67-
**CUDA 12:**
67+
**CUDA 12 (Recommended):**
6868
```bash
6969
# Create and activate conda environment
7070
conda create -n diffct python=3.12
7171
conda activate diffct
7272

73-
# Install CUDA Toolkit, PyTorch, and Numba
73+
# Install CUDA (here 12.8.1 as example) PyTorch, and Numba
7474
conda install nvidia/label/cuda-12.8.1::cuda-toolkit
75-
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
75+
76+
# Install Pytorch, you can find the commend here: https://pytorch.org/get-started/locally/
77+
78+
# Install Numba with CUDA 12
7679
pip install numba-cuda[cu12]
7780

7881
# Install diffct
7982
pip install diffct
8083
```
8184

82-
**CUDA 11:**
85+
<details>
86+
<summary>CUDA 13 Installation</summary>
87+
88+
```bash
89+
# Create and activate conda environment
90+
conda create -n diffct python=3.12
91+
conda activate diffct
92+
93+
# Install CUDA (here 13.0.2 as example) PyTorch, and Numba
94+
conda install nvidia/label/cuda-13.0.2::cuda-toolkit
95+
96+
# Install Pytorch, you can find the commend here: https://pytorch.org/get-started/locally/
97+
98+
# Install Numba with CUDA 13
99+
pip install numba-cuda[cu13]
100+
101+
# Install diffct
102+
pip install diffct
103+
```
104+
105+
</details>
106+
107+
<details>
108+
<summary>CUDA 11 Installation</summary>
109+
83110
```bash
84111
# Create and activate conda environment
85112
conda create -n diffct python=3.12
86113
conda activate diffct
87114

88-
# Install CUDA Toolkit, PyTorch, and Numba
115+
# Install CUDA (here 11.8.0 as example) PyTorch, and Numba
89116
conda install nvidia/label/cuda-11.8.0::cuda-toolkit
90-
pip3 install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118
117+
118+
# Install Pytorch, you can find the commend here: https://pytorch.org/get-started/locally/
119+
120+
# Install Numba with CUDA 11
91121
pip install numba-cuda[cu11]
92122

93123
# Install diffct
94124
pip install diffct
95125
```
96126

127+
</details>
128+
97129
## 📝 Citation
98130

99131
If you use this library in your research, please cite:
100132

101133
```bibtex
102134
@software{diffct2025,
103135
author = {Yipeng Sun},
104-
title = {diffct: Differentiable Computed Tomography
136+
title = {diffct: Differentiable Computed Tomography
105137
Reconstruction with CUDA},
106138
year = 2025,
107139
publisher = {Zenodo},

0 commit comments

Comments
 (0)