When trying to train any classification model (and this is most prevalent for binary classification), the following UserWarning from sklearn pops up at every step per subject: y_pred contains classes not in y_true.
To Reproduce
Steps to reproduce the behavior:
Use any data for binary classification (TIL, EGFR, ...).
Try to train.
See the following warnings:
The ``converters`` are currently experimental. It may not support operations including (but not limited to) Functions in ``torch.nn.functional`` that involved data dimension
[SNIP!]
Starting validation :
C:\Projects\GaNDLF\venv\Lib\site-packages\sklearn\metrics_classification.py:2399: UserWarning: y_pred contains classes not in y_true
warnings.warn("y_pred contains classes not in y_true")
The first warning comes from [this line](https://github.com/M3DV/ACSConv/blob/95dc860a77e309f010a3d8be1f675e77c7dfeda4/acsconv/converters/__init__.py#L6).
### Expected behavior
This warning should come up just once.
### Media
N.A.
### Environment information
<!-- Put the output of the following command:
python ./gandlf_debugInfo
-->
```powershell
GANDLF version: 0.0.20-dev
Git hash: 000ea17ecd2e018dafa2ecfe7a85c030bb4a02cb
Platform: Windows-10-10.0.22631-SP0
Machine: AMD64
Processor: Intel64 Family 6 Model 165 Stepping 2, GenuineIntel
Architecture: 64bit WindowsPE
Python environment:
Version: 3.11.6
Implementation: CPython
Compiler: MSC v.1935 64 bit (AMD64)
Build: main Oct 3 2023 10:29:11
Describe the bug
When trying to train any classification model (and this is most prevalent for binary classification), the following
UserWarning
from sklearn pops up at every step per subject:y_pred contains classes not in y_true
.To Reproduce
Steps to reproduce the behavior:
[SNIP!]
Starting validation :
C:\Projects\GaNDLF\venv\Lib\site-packages\sklearn\metrics_classification.py:2399: UserWarning: y_pred contains classes not in y_true warnings.warn("y_pred contains classes not in y_true")
Additional context
N.A.