Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
Already on GitHub? Sign in to your account
Get NN working a bit better on the tests #16
Merged
Commits
Show all changes
25 commits
Select commit
Hold shift + click to select a range
ecffda8
Added a shell for the nerual net
michaelhush 5f48778
Fix some whitespace errors
charmasaur 326f98b
Fix some minor controller documentation errors
charmasaur 635a5f7
Tweaks to NN learner shell
charmasaur 6a6f663
Remove unnecessary uncertainty stuff from NNL
charmasaur 97d5b23
Added visualization introduced bug
michaelhush e8a8715
NerualNet ready for actually net
michaelhush 2efd317
Fix some NN typos
charmasaur d5c5749
Basic NN learner implementation
charmasaur d7b1fca
Fix number_of_controllers definition
charmasaur 2126150
More NNController tidying/tweaking
charmasaur d78a661
Remove scaler from NNController
charmasaur 34b504b
Tidying/logging for NN impl
charmasaur 9224be5
Fix importing/creation of NN impl
charmasaur f76c9b2
Merge branch 'NeuralNetA' of https://github.com/michaelhush/M-LOOP in…
charmasaur be3c8a5
Pull NNI construction into create_neural_net
charmasaur 3a46a17
Dumb implementation of predict_costs array version
charmasaur 89f1e1a
Set new_params_event in MLC after getting the cost
charmasaur 3e4b3df
Add (trivial) scaler back to NNL
charmasaur f22c979
Don't do one last train in order to predict minima
charmasaur 82fa70a
Merge pull request #15 from charmasaur/NeuralNetA
michaelhush e30906a
Tweak some NNI params to perform better on the test
charmasaur e6d371a
Still print predicted_best_cost even when predicted_best_uncertainty …
charmasaur e6e83e8
Use TF gradient when minimizing NN cost function estimate
charmasaur 1900587
Plot NN surface when there are 2 params
charmasaur
Jump to file or symbol
Failed to load files and symbols.