Skip to content
Snippets Groups Projects
Select Git revision
  • master default protected
1 result
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.015Dec22Oct211430Sep292216Aug23Jul1382123Jun211820May1918115427Apr211486131Mar30292826252423221512111093226Feb25242019181716121095432129Jan2827262522201915141312118723Dec2117161514119427Nov26259Oct52128Sep2523211817161514111098432131Aug28272625242120191817141312111030Jul29282724232221201716151410987632Merge branch 'optim' into ai4ebv-publicmastermasterExample shell script for SLURM sbatch.Merge branch 'optim' into ai4ebv-publicRevert.Merge branch 'optim' into ai4ebv-publicEscape pattern before compiling.Merge branch 'optim' into ai4ebv-publicImplemented support for CyclicalLR scheduling.Implemented support for CyclicalLR scheduling.Merge branch 'optim' into ai4ebv-publicStop training if loss is NaN.Merge branch 'optim' into ai4ebv-publicLog learning rate.Merge branch 'optim' into ai4ebv-publicLog learning rate.Merge branch 'optim' into ai4ebv-publicImplemented LR-scheduling.Public Readme for AI4EBV.Public Readme for AI4EBV.Catch error when coordinates are inverted.Catch error when coordinates are inverted.Change axis order when clipping rasters.Added optional amount of buffering.Added optional amount of buffering.Change clip to gdal>=3.0.Handle negative values for array replacing.Array replace for negative values.Array replace for negative values.Implemented support for multi-label cases.Implemented gradient clipping.Implemented gradient clipping.Remove NaN masking.Remove NaN masking.Moved architecture to climax.Added a strided convolutional downsampling layer.Parameterized saving dataloaders.Added USegNet model.Fix logging of loss.Use correct Early Stopping model.Exclude missing values for regression tasks when computing loss.