ClimbsRocks/data-formatter issues and pull requests

#79 - error with data-formatter

Issue - State: open - Opened by tiblop over 5 years ago

#78 - Error: ValueError: '"continuous"' is not in list

Issue - State: open - Opened by niklaszantner about 8 years ago - 2 comments

#77 - error on data-formatter, to lower function

Issue - State: closed - Opened by MelvinDunn over 8 years ago - 12 comments

#76 - long-term nlp column cleanup

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#75 - update readme for NLP

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#74 - use LabelEncoder on y values

Issue - State: open - Opened by ClimbsRocks over 8 years ago - 3 comments

#73 - delete temporary files

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#72 - horizontally stack the results from tf-idf onto X

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#71 - eventually, add in a sentiment prediction to each nlp field

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#70 - add nlp to featureEngineering

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#69 - update validation.py to allow NLP

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#68 - update API for validationSplit column

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#67 - more obvious input validation logs

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#66 - allow the user to pass in multiple date columns

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#65 - code clean up

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#64 - rename from ppCompleteLocation to machineJSLocation

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#63 - rename kagglePredict to just predict

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#62 - run featureSelecting on only new features in polynomialFeatures

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#61 - move polynomialFeatures to after imputingMissingValues

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#60 - make a module in featureSelecting available widely for easy feature pruning

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#59 - FUTURE: add in a subset of new features, prune out the non-useful ones, repeat

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#58 - add in option to keep all features

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#57 - update feature selecting to select comparatively

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#56 - support using absolute values in feature selecting

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#55 - support groupBy for joined data

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#54 - bug- joined values

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#53 - subtract all dates from each other

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#52 - give the user the ability to run a script doing their own custom data formatting

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#51 - take the log of the regression output column if the range is more than 100

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#50 - update validation.py to handle "date" as an acceptable dataDescription type

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#49 - handle the validation split in df

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#48 - rossman appears to blow up the fileNames test, even though it works on a very large dataset when run outside the test suite

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#47 - adds fileNames test to all our test datasets

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#46 - add an optional flag allowing users to ask for dense matrices

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#45 - write all files as sparse matrices

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#44 - see if we can remove the data files from the git history entirely

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#43 - remove first row

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#42 - make sure we're taking in problemType properly

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#41 - add tests for joining datasets

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#40 - refactor testing datasets into their own module

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#39 - offer a don't prune option

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#38 - switch back to using rfecv

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#37 - more feature engineering

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#36 - look into taking the log of the output column

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#35 - handle having no ID for input data

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#34 - handle dates

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#33 - vectorize output data when relevant

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#32 - get brain.js formatting back up and running

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#31 - reformat for messageParent

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#30 - refactor neural network normalization to work with sparse matrices

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#29 - write pretty names to top of output and id files

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#28 - generally check for methods that copy matrices and delete unnecessary copies

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#27 - check if zip(*matrix) creates a copy of the matrix

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#26 - time for more code cleanup

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 2 comments

#25 - new feature: if possible, add in categorical data as both categorical and continuous

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#24 - new feature: imputed column vs. raw (blank) column

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#23 - add in a new feature: numOfImputedFeatures

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#22 - general code cleanup

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#21 - comment out outputFolder in dfArgs, so that we can test to make sure the output folder is created for us automatically

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#20 - refactor test suite to use index.js and not startPyTest.js

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#19 - post-MVP: allow a flag for brainjs formatting

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#18 - remove brainjs formatting in favor of scikit-neuralnetwork

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#17 - test: add test for collecting fileNames

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#16 - test: add test for invoking callback function with filenames

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#15 - invoke the callback function with an args object that holds the fileNames

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#14 - post-MVP: write ints for categorical columns

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#13 - make sure scientific notation in JSON is ok

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 2 comments

#12 - reduce memory use in basic ways

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#11 - put all python modules into their own folder

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#10 - remove all hard-coded references to 150000

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#9 - create categorical data for testSuite data

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#8 - allow the user to pass in a flag for final output formatting

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#7 - format for brain.js

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#6 - binarize the categorical data

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#5 - feature selection

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#4 - write to file

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

#3 - remove categories that are only present in 1 row

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 2 comments

#2 - summarize data for the user to see

Issue - State: open - Opened by ClimbsRocks over 8 years ago

#1 - POST-MVP make this usable from both the command line and as a node module

Issue - State: closed - Opened by ClimbsRocks over 8 years ago - 1 comment

GitHub / ClimbsRocks/data-formatter issues and pull requests