"
Input: {
Use data for modeling directly
(ftable) - Feature table
[ [f1,f2,...], [f1,f2,...], ...]
(fkeys) - Key names
[ "f1 name", "f2 name", ...]
(ctable) - Outcome to model
[ [value], [value], ... ]
(ckeys) - Key names
[ "value name" ]
OR select entries:
(repo_uoa) or (experiment_repo_uoa) - can be wild cards
(remote_repo_uoa) - if remote access, use this as a remote repo UOA
(module_uoa) or (experiment_module_uoa) - can be wild cards
(data_uoa) or (experiment_data_uoa) - can be wild cards
(repo_uoa_list) - list of repos to search
(module_uoa_list) - list of module to search
(data_uoa_list) - list of data to search
(search_dict) - search dict
(ignore_case) - if 'yes', ignore case when searching
(features_flat_keys_list) - list of flat keys to extract from points into table
(order is important: for example, for plot -> X,Y,Z)
(features_flat_keys_index) - add all flat keys starting from this index
(for example, ##features#)
(characteristics_flat_keys_list) - list of flat keys to extract from points into table
(order is important: for example, for plot -> X,Y,Z)
(characteristics_flat_keys_index) - add all flat keys starting from this index
(for example, ##features#)
Model:
model_module_uoa - model module
model_name - model name
model_file - model file
(keep_temp_files) - if 'yes', keep temp files
(remove_points_with_none) - if 'yes', remote points with None
(model_data_uoa) - record to this data_uoa (container: model_module_uoa)
(model_repo_uoa) - use this repo to record model
}
Output: {
return - return code = 0, if successful
> 0, if error
(error) - error text if return > 0
rmse
prediction_rate
observations
mispredictions
}
"