v0.0.23

• segmentation merge untagged between references (#240)
• seg separate body back figure table default (#239)
• fix ack sub section (#238)
• allow individual fields to be added or removed via cli (#237)
• merge references in segmentation model (#236)
• added alto to text xsl (#233)

v0.0.22

• fixed abstract cut off in segmentation annotator (#227)

v0.0.21

• detect page headers (#222)
• make requirements minimum version (#221)
• moved sciencebeam_gym dependencies to core package (#220)
• fixed html entities after invalid xml (#218)
• upgraded to python 3.7 (#217)
• reduced segmentation entities_by_name log level (#209)

v0.0.20

• improved segmentation model, annotate full text fields, some front matter can appear at the end of the document (#208)

v0.0.19

• disabling expanding to untagged lines by default (#204)

v0.0.18

• figure and table annotator (#188)

v0.0.17

• pull out figures from paragraphs (#182)
• annotate appendix (#181)
• annotate keywords (#180)
• enable fulltext extend to line (#179)
• annotate split paragraph (#178)
• annotate lists within paragraphs (#177)
• annotate list items outside paragraphs as list (#176)
• annotate top level list as paragraph (#175)
• annotate top level body paragraphs (#174)
• improve fulltext figure annotation (#173)
• added auto-annotation for full text model (#169)

v0.0.16

• filter xml on the fly (#167)

v0.0.15

• added mapping for affiliation institution, department and additional address fields (#164)

v0.0.14

• moved annotators to annotation package (#162)
• add affiliation address annotation (#161)