This repository was archived by the owner on Oct 26, 2022. It is now read-only.

Description
When I interactively translate a sentece I get something like this as output:
S-0 ['▁A', '▁black', '▁box', '▁in', '▁your', '▁car', '▁?']
H-0 -0.9591023563900086 ['▁तुम्हारी', '▁गाड़ी', '▁का', '▁काला', '▁डिब्बा', '▁?']
P-0 -1.7602 -0.5362 -1.4817 -1.1891 -1.1348 -0.4620 -0.1497
what is P-0 scores (positional score per token). Are these -ve logarithmic probability(kind of confidence score) ?