georgemilosh/Climate-Learning

Issues

• Compatibility with newer tensorflow versions

  #93 opened 4 months ago by AlessandroLovo
  0

• import params from external folder

  #92 opened 7 months ago by AlessandroLovo
  1

• Move useful unspecific things into separate modules

  #89 opened 7 months ago by AlessandroLovo
  0

• For the gaussian approximation on CESM matrix `W` is 5.7GB

  #90 opened 7 months ago by AlessandroLovo
  1

• Allow to enable and disable mods

  #88 opened 7 months ago by AlessandroLovo
  0

• Put some error message when the year_list exceeds the size of the file we load so that it is clear what is happening

  #60 opened 2 years ago by georgemilosh
  1

• cannot load kernels from composite in IIPR

  #87 opened 8 months ago by AlessandroLovo
  0

• 'monitor' is ignored to compute the score

  #86 opened 8 months ago by AlessandroLovo
  1

• Support for cascading inheritance

  #82 opened 8 months ago by AlessandroLovo
  1

• `probabilistic_regression`: put the activation function for sigma in the model architecture rather than in the loss function

  #85 opened 8 months ago by AlessandroLovo
  0

• It is a bad idea to write to home directory commit #d1ba925cbddd69d00bd297c4cdfc90efc8a5f1f5

  #76 opened 8 months ago by georgemilosh
  0

• pass metrics from command line

  #84 opened 8 months ago by AlessandroLovo
  0

• flexible hypop doesn't remove kwargs at default

  #83 opened 8 months ago by AlessandroLovo
  1

• `lsm2mask` should be handled better

  #81 opened 9 months ago by AlessandroLovo
  0

• CESM France mask could be better

  #79 opened 9 months ago by AlessandroLovo
  6

• ERA_Fields.Plasim_Field.sort_lat doesn't do anything

  #78 opened 9 months ago by AlessandroLovo
  0

• `Learn2_new.py` trainer inheritence

  #71 opened a year ago by georgemilosh
  1

• Transfer Learning from outside folder fails

  #74 opened a year ago by georgemilosh
  11

• dict arguments are not dealt with correctly

  #73 opened a year ago by AlessandroLovo
  1

• Write docstring for gaussian approx

  #63 opened a year ago by AlessandroLovo
  0

• Support for single fold run

  #62 opened 2 years ago by AlessandroLovo
  1

• xarray friendly `ef.create_mask`

  #68 opened a year ago by AlessandroLovo
  2

• Leap years

  #69 opened a year ago by AlessandroLovo
  1

• Queue file

  #72 opened a year ago by AlessandroLovo
  0

• Using `lon` and `lat` as inputs to `X` when classifying heatwaves

  #70 opened a year ago by georgemilosh
  0

• the default behavior should be to load from a repository config file which documents where to look for the data

  #65 opened a year ago by georgemilosh
  8

• The new runs should carry the label `R` before completion

  #67 opened a year ago by georgemilosh
  0

• Reduce memory usage!

  #50 opened 2 years ago by georgemilosh
  12

• Reduce the size of the repository

  #64 opened a year ago by georgemilosh
  0

• transfer learning on `u`

  #53 opened 2 years ago by georgemilosh
  3

• Data not normalized prior to being fed into the model when computing Y_pred_unbiased?

  #58 opened 2 years ago by georgemilosh
  5

• Testing the network before loading data

  #55 opened 2 years ago by georgemilosh
  1

• `setup.sh` could be outdated for the new versions of tensorflow

  #57 opened 2 years ago by georgemilosh
  0

• The error messages return wrong lines for `Learn2_new.py`

  #54 opened 2 years ago by georgemilosh
  5

• seems that this condition is restrictive. I don't see a reason why we need to limit the majority class to never be smaller@georgemilosh

  #52 opened 2 years ago by georgemilosh
  2

• Save the temperature anomaly so we don't need to recompute it

  #38 opened 2 years ago by AlessandroLovo
  2

• `field_infos` should be a json file

  #47 opened 2 years ago by georgemilosh
  1

• Ability to skip convolutional layers and support for weight decay

  #48 opened 2 years ago by georgemilosh
  0

• Working with CESM

  #43 opened 2 years ago by georgemilosh
  0

• Flexible latitude in `committor_projection_NN.py`

  #45 opened 2 years ago by AlessandroLovo
  1

• `nfolds` is kwarg of both `prepare_XY` and `k_fold_cross_val` which can cause problems

  #46 opened 2 years ago by georgemilosh
  2

• Merge the subm2 and main

  #39 opened 2 years ago by georgemilosh
  1

• Make more readable and compact notebooks

  #37 opened 2 years ago by AlessandroLovo
  0

• Implement analog method in VAE

  #36 opened 2 years ago by georgemilosh
  0

• Better exploitation of xarray features

  #31 opened 2 years ago by AlessandroLovo
  1

• Use submodules

  #34 opened 2 years ago by AlessandroLovo
  2

• More information in README.md

  #35 opened 2 years ago by AlessandroLovo
  0

• Enable the possibility to run autoencoder on reduced space using main branch

  #33 opened 2 years ago by georgemilosh
  0

• Better masked average

  #32 opened 2 years ago by AlessandroLovo
  1

• More flexible choice of the data domain

  #30 opened 2 years ago by AlessandroLovo
  1