Tesseract LSTM training info

[k.elta...@gmail.com]

During training some info are written to stdout but I don't understand them can any one help?

At iteration 3053/40000/40054, Mean rms=0.25%, delta=0.461%, char train=1.501%, word train=2.578%, skip ratio=0.2%,  wrote checkpoint.

what is 3053/40000/40054?

Loaded 83/1299 pages (2-84) of document train/features/this/features/eng.99903898.DejaVu_Sans_Mono_Bold.exp-2.lstmf

what is 83/1299 & (2-84)?

Also what is the iteration size? so does one iteration implies going over the whole data or what?