Skip to content
Snippets Groups Projects
Select Git revision
  • master default
1 result
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.04Dec29Nov272216131230Oct26229826Jul23201927Jun262521Apr929Mar28261327Feb1930Jan2213Dec1229Nov16expe keras subsample conv handmastermasterremove keras evaluation outputcontinue setting up experiment keras end to endSetup keras experiment end_to_endFix issues deepfriedconvnet and mnistremove this useless lazyfileexpe lazyfile lazyfile_classif_end_to_end_with_augment_solve_mnist_deepfried_plus_rbfExpe end to end with augment without dense (problem cluster)expe classif_end_to_end_with_augment_solve_mnist_deepfried lazyfileadd reports to the ignored part of the projectgraph drawing classif_end_to_end_subsample_conv_hand_with_augment with mixed resultschange in tensorboard path in end_to_end_subsample_conv_handgraph drawing classif_end_to_end_with_augmentinit classif_end_to_end_subsample_conv_hand_with_augment graph drawing + init classif_end_to_end_with_augment + add some tensorboard logging in end_to_end_with_2_layers_only_dense_with_augmenttensorboard file with kernel nameexpe /end_to_end_subsample_conv_hand_with_augment script + lazyfileexp end_to_end_subsample_conv_hand_steph_model maj du fichier de log tensorboard + do not recompute repr after trainingsolve infinite loop issueinit expe end_to_end_subsample_conv_hand_steph_model scriptexpe end_to_end_with_augment script + lazyfileinit expe /end_to_end_with_augment scriptmore tensorboard logging + reset default graphupgrade tensorboard optionsolve issue count batchexp classif_end_to_end_subsample_conv_hand graphexp script end_to_end_with_2_layers_only_dense_with_augmentexp create script end_to_end_2_layers_only_dense_with_augmentexp lazyfile_classif_end_to_end_subsample_conv_hand script + paramexpe classif_end_with_2_layers_only_dense script + lazyfile + graph_drawingadd all expe files for until october 2018CLuger now create output folder if it doesn't existfew_data_batchnorm graph drawing scriptchange archi so that scripts/parameterfiles/graphdrawing looks like each othersolve importsnow uses deepstromlayer class + tensorboard supportupdate gitignore to not take files with debug in the namemove graph drawing by date + remove some useless scriptsrename benchmark_vgg to benchmark_classificationclean benchmark_vggupdate multiview script: polynomial kernel added, new support to transformers