API Quick referenceΒΆ

Data item HTTP method Resource What? Description
Dataset POST dataset Upload a dataset First step
  GET dataset Download a cleaned dataset  
  DELETE dataset Delete a dataset All associated analyses and sub analyses are deleted
  GET datasets List of existing datasets  
Analysis POST analysis Create analysis Start a new analysis on the uploaded dataset
  GET analysis Get a list of all sub analyses  
  DELETE analysis Delete analysis All sub analyses are deleted
Sub analysis POST sub_analysis Create a sub analysis Start a new sub analysis
  DELETE sub_analysis Delete a sub analysis Discard this sub analysis. Frees up space
bins POST target_bins Define target bins For numerical target attributes, user has to specify bins (value ranges)
  GET target_bins Get target bins  
attributes POST attributes_detail Update attribute information Use this call to override column labels, specify ignore columns etc
  GET attributes_detail Get all attributes and their properties  
  GET iclean Initiate data cleansing Returns attributes summary and cleansing summary. Cleaned data file is not returned by this call. See Download Dataset
sub analysis GET fastfill Imputation Auto fill missing values
  GET automerge Merge nultiple levels Merge levels with similar correlation to target class levels
  GET learning_curve Learning curves May be Good, Inadequate or inconclusive
  GET dNoise Merge & Bin data Remove the noise
  GET nReduce Dimensionality reduction Feature Selection
  GET oDetect Outlier detection Too Good To be True (TGTBT)
  POST set_attr_importance Mark important attributes User overrides
sub analysis POST irule Rule extraction Extract hidden insights
  POST act_rule WhatIf analysis Returns the insights that would match the given desired class level
  POST generalize_rule Remove one attribuet condition in RULE  
  POST specialize_rule Add an additional condition  
  POST customize_rule Change value of one attribute in a rule condition  
  POST save_rule Save the generated rule Save the rule generated by act/generalize/specialize/customize rule operation (Yet to implement)
sub analysis POST ilearn Generate model Basing on the error metric, model with best performance is generated
  POST predictions Run predictions on the evaluation data Uses the Model that was trained in the iLearn
prediction GET prediction Get prediction data To see the predictions. Generate predictions first
  POST create_advisor_model Generate advisor model  
  POST iadvsie Generate predictions on the target Generates predictions on the target for the uploaded Eval data using the Model generated by iLearn
  GET iadvsie Get suggestions Get suggestions based on suggestion id
  POST iadvise_with_predict Generate predictions on the target when eval data has prediction Similar to iadvise, but when uploaded eval data has prediction