Soma

Sonal G. any thoughts on this??

yes but how will we take the call that this amount of training is enough. Because the current approach we follow is do initial ground truth training and then over time mature ground truth file with more rows and re-train the zing (training method) and do predict for improvised clustering . One question we have is as and when we have more ground truth identified by human do we need to re-train again and again?

hi Sonal G. I also see from documentation incremental processing is possible but I have a curious question is incremental training is available in enterprise version, let us say first I have zingg trained using 100 rows and lets say I have identified few more training (ground truth records -around 1K records) . Is it possible that we can do incremental training or should we re-train for all ?

Ajit k. can u please let me know the DBR and Zingg Jar you have used because I am getting errors running zingg on databricks personal compute

thanks Sonal G. for helping out

also final question is there any doc where we can have DBR Zingg jar Compatability

hi Sonal G., thanks,also assume we have already some ground truth with https://docs.zingg.ai/latest/stepbystep/createtrainingdata/addowntrainingdata , can we directly use match phase or should we use train and match phases even while having training data and our idea is we will accumulate the trainign data (ground truth) with more and more rows and do entity Resolution for eg (100 rows) -> 1000K ( zingg identifies and added to groundtruth) and do ER on 100K records and so on , So if we can use match alone?

hi Sonal G. thanks for the response I have a query so should we read as pandas dataframe or can you share some example where delta table is read . Because in examples I see only CSVPipe option which I am not sure if it uses spark dataframe or pandas dataframe