<div>hi alex,</div><div><br>
> (0) [1 (2) ^ aj-hd_scp_c av_-_s-cp-mc-pr_le "well"] 0.522613 {36392 447 36392 445} [0 1]<br>
<br>
</div><div dir="auto">in a nutshell, the format is</div><div dir="auto"><br></div><div dir="auto">(id) [tid (parameter*) symbol] weight {counts} [range]</div><div dir="auto"><br></div><div dir="auto">the feature identiers are unique consecutive integers.</div><div dir="auto"><br></div><div dir="auto">template identifiers are integers, specified through comments in the top of ’features.lisp‘ in [incr tsdb()].</div><div dir="auto"><br></div><div dir="auto">each template can take zero to two parameters, e.g. the grandparenting level in type-1 features; see the above file and zhang, oepen, & carroll (2007; IWPT).</div><div dir="auto"><br></div><div dir="auto">symbol is the representation of the feature proper, again interpreted specific to the template type, e.g. two grandparents in the above (of which one is the tree ’top sentinal‘), then the mother, then the sequence of daughters (singleton, in this case).</div><div dir="auto"><br></div><div dir="auto">the counts and (minimum and maximum) value range information is there for bookkeeping purposes; MEM scoring does not depend on this.</div><div dir="auto"><br></div><div dir="auto">MEM training is supported in [incr tsdb()] in the LOGON tree; documented to some degree on the wiki, i believe. as of this summer, i think woodley offers an alternative full-forest trainer. i recommend you look into that!</div><div dir="auto"><br></div><div dir="auto">best wishes, oe</div><div dir="auto"><br></div>