DATA FRORMAT

METADATA

id - the sentence ID according to the Latvian UD Treebank data.
snt_lv - the original sentence (Latvian).
snt - the machine-translated sentence (English).
alignments_translation - token-level alignment between the Latvian and English sentences.
alignments - alignment between tokens (English) and nodes of the AMR graph (see the alignment format: https://github.com/lil-lab/amr/blob/07b581e7d160fa2625eeefa86ae5e9fe5c589be2/utils/jamr/docs/Alignment_Format.md).

AMR GRAPH

The English translation encoded as an AMR graph in the PENMAN notation (see the language specification: https://amr.isi.edu/language.html).

Each record in the dataset is separated by an empty line. The dataset is split into subsets for training, development and testing using the same split as for the Latvian UD Treebank (v2.5).

EXAMPLE

# ::id a-p3315-p3s6
# ::snt Sister Ilze is also a model .
# ::snt_lv Māsa Ilze arī ir modelētāja.
# ::alignments 0-1|0.2+0.2.0+0.2.0.0 1-2|0.1+0.1.0+0.1.0.0 3-4|0.0 5-6|0
# ::alignments_translation 0-0 1-1 3-2 2-3 4-4 4-5 5-6
(v7 / modelētāja                           		-> 0
    :mod (v6 / also)                        		-> 0.0
    :domain (v4 / person                    		-> 0.1
        :name (v5 / name                    		-> 0.1.0
            :op1 "Ilze"))                    		-> 0.1.0.0
    :poss (v1 / person                      		-> 0.2
        :ARG0-of (v2 / have-rel-role-91     		-> 0.2.0
            :ARG2 (v3 / sister))))          		-> 0.2.0.0
Sister: (0-1) => v1+v2+v3 (0.2+0.2.0+0.2.0.0)
Ilze: (1-2) => v4+v5+op1+wiki (0.1+0.1.0+0.1.0.0)
also: (3-4) => v6 (0.0)
model: (5-6) => v7