@@ -43,7 +43,7 @@ in a cell will verify this has worked and show you what kind of hardware you hav
43
43
44
44
#### Google Colab Setup (CUDA 12.x, PyTorch 2.6, MONAI 1.5)
45
45
46
- In Google Colab, the default environment may cause version conflicts with MONAI.
46
+ In Google Colab, the default environment may cause version conflicts with MONAI.
47
47
To ensure compatibility, install PyTorch and MONAI explicitly as follows:
48
48
49
49
# Install PyTorch 2.6.0 with CUDA 12.4
@@ -56,19 +56,19 @@ pip install "monai[all]" nibabel pydicom ipywidgets==8.1.2
56
56
57
57
### Known issues and fixes
58
58
59
- - Torchaudio mismatch
60
- Colab may come with torchaudio 2.8.0, which is incompatible with torch 2.6.0.
59
+ - Torchaudio mismatch
60
+ Colab may come with torchaudio 2.8.0, which is incompatible with torch 2.6.0.
61
61
Installing the versions above resolves this issue.
62
62
63
- - filelock conflicts with nni
64
- Some preinstalled packages (such as pytensor with newer filelock) may conflict.
63
+ - filelock conflicts with nni
64
+ Some preinstalled packages (such as pytensor with newer filelock) may conflict.
65
65
Use the following commands to fix:
66
66
67
67
pip uninstall -y pytensor
68
68
pip install -U filelock
69
69
70
- - Too many workers warning
71
- Colab has limited CPU resources, and high num_workers settings may freeze execution.
70
+ - Too many workers warning
71
+ Colab has limited CPU resources, and high num_workers settings may freeze execution.
72
72
It is recommended to use --num_workers=2 when running tutorials.
73
73
74
74
0 commit comments