georgemilosh/Climate-Learning issues and pull requests

#91 - 90

Pull Request - State: closed - Opened by AlessandroLovo about 1 year ago

#90 - For the gaussian approximation on CESM matrix `W` is 5.7GB

Issue - State: closed - Opened by AlessandroLovo about 1 year ago - 1 comment
Labels: enhancement

#89 - Move useful unspecific things into separate modules

Issue - State: open - Opened by AlessandroLovo about 1 year ago
Labels: enhancement, question

#88 - Allow to enable and disable mods

Issue - State: closed - Opened by AlessandroLovo about 1 year ago
Labels: enhancement

#87 - cannot load kernels from composite in IIPR

Issue - State: open - Opened by AlessandroLovo about 1 year ago
Labels: bug

#86 - 'monitor' is ignored to compute the score

Issue - State: open - Opened by AlessandroLovo about 1 year ago - 1 comment
Labels: bug

#85 - `probabilistic_regression`: put the activation function for sigma in the model architecture rather than in the loss function

Issue - State: closed - Opened by AlessandroLovo about 1 year ago
Labels: bug

#84 - pass metrics from command line

Issue - State: open - Opened by AlessandroLovo about 1 year ago
Labels: enhancement

#83 - flexible hypop doesn't remove kwargs at default

Issue - State: closed - Opened by AlessandroLovo about 1 year ago - 1 comment
Labels: bug

#82 - Support for cascading inheritance

Issue - State: closed - Opened by AlessandroLovo about 1 year ago - 1 comment
Labels: enhancement

#81 - `lsm2mask` should be handled better

Issue - State: open - Opened by AlessandroLovo about 1 year ago
Labels: bug

#80 - Cesm xarray france

Pull Request - State: closed - Opened by AlessandroLovo about 1 year ago - 2 comments

#79 - CESM France mask could be better

Issue - State: closed - Opened by AlessandroLovo about 1 year ago - 6 comments
Labels: enhancement

#78 - ERA_Fields.Plasim_Field.sort_lat doesn't do anything

Issue - State: open - Opened by AlessandroLovo about 1 year ago
Labels: bug

#77 - Fix typos

Pull Request - State: closed - Opened by omahs over 1 year ago

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

Issue - State: closed - Opened by georgemilosh over 1 year ago
Labels: enhancement, help wanted, invalid

#75 - #74

Pull Request - State: closed - Opened by AlessandroLovo over 1 year ago - 1 comment
Labels: enhancement

#74 - Transfer Learning from outside folder fails

Issue - State: closed - Opened by georgemilosh over 1 year ago - 11 comments
Labels: bug, enhancement, help wanted

#73 - dict arguments are not dealt with correctly

Issue - State: closed - Opened by AlessandroLovo over 1 year ago - 1 comment
Labels: bug

#72 - Queue file

Issue - State: open - Opened by AlessandroLovo over 1 year ago
Labels: enhancement

#71 - `Learn2_new.py` trainer inheritence

Issue - State: open - Opened by georgemilosh over 1 year ago - 1 comment

#70 - Using `lon` and `lat` as inputs to `X` when classifying heatwaves

Issue - State: open - Opened by georgemilosh over 1 year ago
Labels: enhancement, wontfix

#69 - Leap years

Issue - State: open - Opened by AlessandroLovo over 1 year ago - 1 comment
Labels: bug, enhancement

#68 - xarray friendly `ef.create_mask`

Issue - State: open - Opened by AlessandroLovo almost 2 years ago - 2 comments
Labels: enhancement

#67 - The new runs should carry the label `R` before completion

Issue - State: closed - Opened by georgemilosh almost 2 years ago
Labels: enhancement, help wanted

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

Issue - State: open - Opened by georgemilosh almost 2 years ago - 8 comments
Labels: enhancement

#64 - Reduce the size of the repository

Issue - State: open - Opened by georgemilosh almost 2 years ago
Labels: enhancement, wontfix

#63 - Write docstring for gaussian approx

Issue - State: closed - Opened by AlessandroLovo almost 2 years ago
Labels: documentation

#62 - Support for single fold run

Issue - State: open - Opened by AlessandroLovo almost 2 years ago - 1 comment
Labels: enhancement

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

Issue - State: open - Opened by georgemilosh almost 2 years ago - 1 comment
Labels: invalid

#58 - Data not normalized prior to being fed into the model when computing Y_pred_unbiased?

Issue - State: closed - Opened by georgemilosh almost 2 years ago - 5 comments
Labels: bug, invalid

#57 - `setup.sh` could be outdated for the new versions of tensorflow

Issue - State: closed - Opened by georgemilosh about 2 years ago
Labels: documentation, enhancement

#55 - Testing the network before loading data

Issue - State: open - Opened by georgemilosh about 2 years ago - 1 comment
Labels: enhancement

#54 - The error messages return wrong lines for `Learn2_new.py`

Issue - State: closed - Opened by georgemilosh about 2 years ago - 5 comments
Labels: bug

#53 - transfer learning on `u`

Issue - State: open - Opened by georgemilosh about 2 years ago - 3 comments
Labels: enhancement, invalid, wontfix

#52 - 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

Issue - State: open - Opened by georgemilosh about 2 years ago - 2 comments
Labels: enhancement, wontfix

#50 - Reduce memory usage!

Issue - State: open - Opened by georgemilosh about 2 years ago - 12 comments
Labels: enhancement

#48 - Ability to skip convolutional layers and support for weight decay

Issue - State: closed - Opened by georgemilosh about 2 years ago
Labels: enhancement

#47 - `field_infos` should be a json file

Issue - State: closed - Opened by georgemilosh about 2 years ago - 1 comment
Labels: enhancement

#46 - `nfolds` is kwarg of both `prepare_XY` and `k_fold_cross_val` which can cause problems

Issue - State: closed - Opened by georgemilosh about 2 years ago - 2 comments
Labels: enhancement, wontfix

#45 - Flexible latitude in `committor_projection_NN.py`

Issue - State: closed - Opened by AlessandroLovo about 2 years ago - 1 comment
Labels: enhancement

#44 - The ability to work with CESM data using ERA_Fields_New.py and Learn2…

Pull Request - State: closed - Opened by georgemilosh about 2 years ago - 8 comments
Labels: enhancement

#43 - Working with CESM

Issue - State: closed - Opened by georgemilosh about 2 years ago
Labels: enhancement

#41 - Subm22main

Pull Request - State: closed - Opened by georgemilosh about 2 years ago

#40 - Subm2 -:> main

Pull Request - State: closed - Opened by georgemilosh about 2 years ago
Labels: enhancement

#39 - Merge the subm2 and main

Issue - State: closed - Opened by georgemilosh about 2 years ago - 1 comment
Labels: enhancement

#38 - Save the temperature anomaly so we don't need to recompute it

Issue - State: closed - Opened by AlessandroLovo over 2 years ago - 2 comments
Labels: enhancement

#37 - Make more readable and compact notebooks

Issue - State: open - Opened by AlessandroLovo over 2 years ago
Labels: documentation, enhancement

#36 - Implement analog method in VAE

Issue - State: closed - Opened by georgemilosh over 2 years ago

#35 - More information in README.md

Issue - State: open - Opened by AlessandroLovo over 2 years ago
Labels: documentation

#34 - Use submodules

Issue - State: closed - Opened by AlessandroLovo over 2 years ago - 2 comments
Labels: enhancement

#33 - Enable the possibility to run autoencoder on reduced space using main branch

Issue - State: closed - Opened by georgemilosh almost 3 years ago

#32 - Better masked average

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment
Labels: enhancement

#31 - Better exploitation of xarray features

Issue - State: open - Opened by AlessandroLovo almost 3 years ago - 1 comment
Labels: enhancement

#30 - More flexible choice of the data domain

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#29 - xarray adaptations in Learn2_New

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#28 - xarray masks

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#27 - xarray Plasim Field

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#26 - Fix the bug with scheduler_kwargs not showing up in config file

Issue - State: closed - Opened by georgemilosh almost 3 years ago - 2 comments
Labels: bug

#25 - Try subtracting the average yearly temperature for computing the labels

Issue - State: open - Opened by georgemilosh almost 3 years ago - 2 comments

#24 - Support for padding that respects the topology of the sphere

Issue - State: open - Opened by georgemilosh almost 3 years ago
Labels: enhancement

#23 - Change the skip connection structure from dictionary to a list

Issue - State: closed - Opened by georgemilosh almost 3 years ago

#22 - Add possibility to train a model without t2m field that is used as a label

Issue - State: closed - Opened by georgemilosh almost 3 years ago

#21 - Add possibility of continuing runs

Issue - State: open - Opened by AlessandroLovo almost 3 years ago
Labels: enhancement

#20 - Accelerate the VAE routines

Issue - State: closed - Opened by georgemilosh almost 3 years ago

#19 - mrso masking bug

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 4 comments
Labels: bug

#18 - Restrict labels to heat wave onsets

Issue - State: open - Opened by georgemilosh almost 3 years ago - 2 comments
Labels: enhancement, wontfix

#17 - early stopping for eons

Issue - State: open - Opened by AlessandroLovo almost 3 years ago
Labels: enhancement, wontfix

#16 - Transfer learning for resampling

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#15 - More modular transfer learning

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 1 comment

#14 - Fix errors with resampling.py

Issue - State: closed - Opened by georgemilosh almost 3 years ago - 1 comment

#13 - Use Learn2_new.py for vae_learn2.py

Issue - State: closed - Opened by georgemilosh almost 3 years ago

#12 - Multiclass classification

Issue - State: open - Opened by AlessandroLovo almost 3 years ago
Labels: enhancement, wontfix

#11 - Sign of tau

Issue - State: closed - Opened by AlessandroLovo almost 3 years ago - 2 comments

#10 - Proper plotting routines for rolled arrays

Issue - State: closed - Opened by georgemilosh about 3 years ago - 2 comments

#9 - Xarrays

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 2 comments

#8 - Better transfer learning

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#7 - Early stopping

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#6 - Discrete metric computed on the fly don't unbias the probabilities

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 2 comments

#5 - utilities

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#4 - Efficient computations over multiple runs

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#3 - runs.json

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#2 - Better logging

Issue - State: closed - Opened by AlessandroLovo about 3 years ago - 1 comment

#1 - :sparkles: Modular support for PLASIM/Learn2.py :sparkles:

Issue - State: closed - Opened by georgemilosh about 3 years ago - 2 comments

GitHub / georgemilosh/Climate-Learning issues and pull requests