Skip to content
Snippets Groups Projects
Select Git revision
  • dev default
  • master protected
2 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.011Feb7328Jan22Dec20329Nov221312876430Oct252416151411874224Sep20181076527Aug129876230Jul2625Jun20730May23222120171614109872130Apr29242215Mar1429Feb22765130Jan2323Dec22211915628Nov24231030Oct262019127630Aug291728Jul2019181711330Jun23Mar17141018Feb175Dec16Nov23Sep1326Aug19Jul17Jun27May2423Merge branch 'dev' of https://git.ecdf.ed.ac.uk/swain-lab/wela into devdevdevtweakfeature(examples): added growth_rate_from_pkl.pyfeature(dataloader): added drop_nan_rows as optionchange(run_wela): changed to GFP_Zfix: typo in run_wela.pychange(run_wela): tpt_end -> tp_endchange(imageviewer): tpt->tp; initial_tpfix(dataloader): error in position, trap, cellfix(growth_rate): Inf causing errorsmastermastersplit notebooks and tried out lstm.feature(imageviewer): highlight a single cellfix(dataloader): id's trap and cell ordered wrongfix(plot_lineage): failed if only one subplotchange(examples): GFP_Z to run_welaminimal feature LSTM built, result is not very good yetvisualise evaluation of imputationminimal feature LSTM model built (without hyperparam finetuning and with minimal training - 10 epochs), saved as lstm_model.keras;added comments about how evaluation for the imputation methods are done.Merge branch 'master' of https://git.ecdf.ed.ac.uk/swain-lab/welachange(bud_to_bud): added label variablerefactor(LSTM_w_MI): remove empty code cell from Jupyter notebookbenchmarked different SVM parameters and ways of gridsearch; benchmarked different imputation methods and their evaluation, thus finalising optimal way to run MI codes.uploaded data files for mutual information estimation; uploaded data for processing some data thru aliby and welacommitting jupyter notebooks for optimising SVM part of MI estimation, and using LSTM in place of SVM.Merge remote-tracking branch 'origin/HEAD'feature(dataloader): check postprocessing in h5sfix(parse_h5): removed spurious breakpointcorrected if statements to avoid ambiguity of CrangeUpdate README.mdchange(run_MI): added axes labelsfix(run_MI): fixed importfeature: added MIdecoding and an example scriptchange(plot_binned): variables renamed for clarityfeature(dataloader): add_array_to_dffeature(get_time_series): returns dataloader IDsfix(plot_lineage): shading for all subplotschange: include_whole_cell -> add_whole_cellMerge branch 'master' of https://git.ecdf.ed.ac.uk/swain-lab/welafix(flavin_filter): now skips all nan signal