Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / ClimbsRocks/ensembler issues and pull requests
#30 - write testDataPredictions and testIDColumn to file
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#29 - add in idHeader and outputHeader to the final file
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#28 - add in the matrixOutput capability again
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#27 - investigate why we're making dual predictions for some multi-category rows
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#26 - add the IDs back into the final output
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#25 - go through and undo the label encoding if applicable
Issue -
State: closed - Opened by ClimbsRocks over 8 years ago
- 1 comment
#24 - binarize the output predictions for multiclass
Issue -
State: open - Opened by ClimbsRocks over 8 years ago
- 2 comments
#23 - enable probabilities for multiclass output
Issue -
State: open - Opened by ClimbsRocks over 8 years ago
#22 - clean up utils.js- do not pass in callback functions
Issue -
State: open - Opened by ClimbsRocks almost 9 years ago
#21 - clean up utils.js- remove args parameter and just make it global
Issue -
State: open - Opened by ClimbsRocks almost 9 years ago
#20 - pull in matrixOutput logic from makePredictions
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#19 - we are not currently invoking machineJS when --ensemble is passed in to machineJS
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#18 - make sure this works with multi-category predictions
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#17 - print out summary of round 0 predictions, in addition to summary of final round predictions
Issue -
State: open - Opened by ClimbsRocks almost 9 years ago
#16 - sort down to only the top 5 most accurate prediction columns
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#15 - rename from ppCompleteLocation to machineJSLocation
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#14 - refactor reading in files
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#13 - adjust numRounds and numIterationsPerRound for machineJS's ensembling round
Issue -
State: closed - Opened by ClimbsRocks almost 9 years ago
- 1 comment
#12 - make sure we're aggregating stage 0 predictions correctly
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#11 - take in the ensembledPredictions folder for the final stage
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#10 - bug: different validation lengths
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#9 - bug: reading in files slows down big time
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#8 - how to ensemble
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#7 - time to ensemble!!
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#6 - consider penalizing algos with a big negative delta between trainScore and validationScore
Issue -
State: open - Opened by ClimbsRocks about 9 years ago
#5 - calculate correlations and variances and means and modes and medians and outliers and such for the set of all predictions
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#4 - ensembling methods
Issue -
State: open - Opened by ClimbsRocks about 9 years ago
- 1 comment
#3 - take in pretty names for output file
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#2 - take in flag for binary output
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment
#1 - remove globalArgs
Issue -
State: closed - Opened by ClimbsRocks about 9 years ago
- 1 comment