Imad Toubal
Imad Toubal
## Maintenance request summary `Inferer.__call__()` function in solaris.nets.infer.py does not properly use torch.no_grad(). PyTorch is effectively computing gradient for the inference process, thus, using significantly more GPU memory. ## Task...
Added `requirements.txt`
Fixed the issue where the function always created a model with 3 input channels and 1 output channel regardless of the `config` object values for `img_ch` and `output_ch`
Currently, the following exception is raised when the uploaded image is a 4-channel png. ``` RuntimeError: The size of tensor a (4) must match the size of tensor b (3)...
Hello, I have downloaded the Webvision 2.0 dataset from https://data.vision.ee.ethz.ch/cvl/webvision/download.html but I don't seem to find any download links for the metadata associated with the validation or test sets. i.e.,...
Refactors include improved readibility in softnms_pytorch.py and consistent use of pytorch to improve performance.
How do we cite this work?