Hi Prashant,
The total document count is not as important as the number of documents per layout trained. Decipher should be capable of recognising most data from a single layout within a handful of examples, if not, then you may need to make changes to your DFD to help. Changes such as additional sample headers, Format Expressions and Validation Lists.
If you have done a lot of training that has been unsuccessful, it can be useful to start again by deleting your training data. It's not 100% necessary, but it can help speed up the new round of training. If you have the latest version (1.2), you can export your training data for safe keeping.
Thanks
------------------------------
Ben Lyons
Product Consultant
Blue Prism
UK
------------------------------
Ben Lyons
Principal Product Specialist - Decipher
SS&C Blue Prism
UK based