autoencodix 0.3.0__tar.gz → 0.3.1__tar.gz

{autoencodix-0.3.0 → autoencodix-0.3.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.5
 Name: autoencodix
-Version: 0.3.0
+Version: 0.3.1
 Summary: Framework for multi-omics data integration by autoencoders.
 Project-URL: Homepage, https://github.com/jan-forest/autoencodix_package
 Project-URL: Issues, https://github.com/jan-forest/autoencodix_package/issues
@@ -378,6 +378,22 @@ export CUBLAS_WORKSPACE_CONFIG=:16:8
 ```
 </details>
 
+<details>
+<summary><b>Python crashes with a segmentation fault as soon as training starts</b></summary>
+
+If the Python process or Jupyter kernel dies without a Python error when you call `fit()` or `run()`, the installed PyTorch build may not work with your GPU driver. We have seen this with `torch 2.15.0` (CUDA 13.2 build) under WSL2. The crash happens when PyTorch initializes CUDA and also occurs with `device="cpu"`. You can confirm it with:
+
+```bash
+python -c "import torch; print(torch.__version__, torch.cuda.is_available())"
+```
+
+If this command crashes as well, install an older PyTorch version:
+
+```bash
+pip install "torch<2.15"
+```
+</details>
+
 <details>
 <summary><b>Which architecture should I choose?</b></summary>
 

{autoencodix-0.3.0 → autoencodix-0.3.1}/README.md

@@ -329,6 +329,22 @@ export CUBLAS_WORKSPACE_CONFIG=:16:8
 ```
 </details>
 
+<details>
+<summary><b>Python crashes with a segmentation fault as soon as training starts</b></summary>
+
+If the Python process or Jupyter kernel dies without a Python error when you call `fit()` or `run()`, the installed PyTorch build may not work with your GPU driver. We have seen this with `torch 2.15.0` (CUDA 13.2 build) under WSL2. The crash happens when PyTorch initializes CUDA and also occurs with `device="cpu"`. You can confirm it with:
+
+```bash
+python -c "import torch; print(torch.__version__, torch.cuda.is_available())"
+```
+
+If this command crashes as well, install an older PyTorch version:
+
+```bash
+pip install "torch<2.15"
+```
+</details>
+
 <details>
 <summary><b>Which architecture should I choose?</b></summary>
 

{autoencodix-0.3.0 → autoencodix-0.3.1}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "autoencodix"
-version = "0.3.0"
+version = "0.3.1"
 authors = [
   { name="Maximilian Joas & Jan Ewald", email="jan.ewald@uni-leipzig.de" },
 ]