From e4e93947390e98e09ce20e346898cf365e01084a Mon Sep 17 00:00:00 2001 From: Franck Dary <franck.dary@lis-lab.fr> Date: Tue, 3 Dec 2019 21:32:07 +0100 Subject: [PATCH] Improved features models --- .../lemmatizer_rules_incremental.fm | 22 +++++++++++++++++++ .../data/feature_models/morpho_incremental.fm | 16 ++++++++++++++ .../data/feature_models/tagger_incremental.fm | 22 ++++++++++++++++--- .../feature_models/tokenizer_incremental.fm | 13 +++++++++++ 4 files changed, 70 insertions(+), 3 deletions(-) create mode 100644 UD_any/data/feature_models/lemmatizer_rules_incremental.fm diff --git a/UD_any/data/feature_models/lemmatizer_rules_incremental.fm b/UD_any/data/feature_models/lemmatizer_rules_incremental.fm new file mode 100644 index 0000000..4606e3b --- /dev/null +++ b/UD_any/data/feature_models/lemmatizer_rules_incremental.fm @@ -0,0 +1,22 @@ +# FORM +b.-1#FORM +b.0#FORM +# POS +b.-1#POS +b.0#POS +# MORPHO +b.-1#MORPHO +b.0#MORPHO +# Suffixes +b.0#FORM.PART.-4.-4 +b.0#FORM.PART.-3.-3 +b.0#FORM.PART.-2.-2 +b.0#FORM.PART.-1.-1 +# Prefixes +b.0#FORM.PART.0.0 +b.0#FORM.PART.1.1 +b.0#FORM.PART.2.2 +b.0#FORM.PART.3.3 +# EOS +b.-1#EOS +b.-2#EOS diff --git a/UD_any/data/feature_models/morpho_incremental.fm b/UD_any/data/feature_models/morpho_incremental.fm index 2e0ecc9..9639583 100644 --- a/UD_any/data/feature_models/morpho_incremental.fm +++ b/UD_any/data/feature_models/morpho_incremental.fm @@ -32,3 +32,19 @@ raw.3 raw.4 raw.5 raw.6 +# Features methode incrementale +s.0#FORM.fasttext +s.0.ldep#FORM.fasttext +s.0#POS +s.0.ldep#POS +s.0#MORPHO +s.0.ldep#MORPHO +s.0.ldep.ldep#MORPHO +s.0.ldep#LABEL +s.0#DIST.s.1 +s.0#nbl +s.0#LABEL +b.0#LABEL +# EOS +b.-1#EOS +b.-2#EOS diff --git a/UD_any/data/feature_models/tagger_incremental.fm b/UD_any/data/feature_models/tagger_incremental.fm index 50ff0fc..a28719c 100644 --- a/UD_any/data/feature_models/tagger_incremental.fm +++ b/UD_any/data/feature_models/tagger_incremental.fm @@ -20,9 +20,6 @@ b.0#FORM.PART.0.0 b.0#FORM.PART.1.1 b.0#FORM.PART.2.2 b.0#FORM.PART.3.3 -# MORPHO -b.-2#MORPHO -b.-1#MORPHO # RAW INPUT raw.0 raw.2 @@ -30,3 +27,22 @@ raw.3 raw.4 raw.5 raw.6 +# Features methode incrementale +s.0#FORM.fasttext +s.0.ldep#FORM.fasttext +s.0#POS +s.0.ldep#POS +s.0#MORPHO +s.0.ldep#MORPHO +s.0.ldep.ldep#MORPHO +s.0.ldep#LABEL +s.0#DIST.s.1 +s.0#nbl +s.0#LABEL +b.0#LABEL +# MORPHO +b.-2#MORPHO +b.-1#MORPHO +# EOS +b.-1#EOS +b.-2#EOS diff --git a/UD_any/data/feature_models/tokenizer_incremental.fm b/UD_any/data/feature_models/tokenizer_incremental.fm index d10a62e..06d3e51 100644 --- a/UD_any/data/feature_models/tokenizer_incremental.fm +++ b/UD_any/data/feature_models/tokenizer_incremental.fm @@ -29,6 +29,19 @@ raw.3 raw.4 raw.5 raw.6 +# Features methode incrementale +s.0#FORM.fasttext +s.0.ldep#FORM.fasttext +s.0#POS +s.0.ldep#POS +s.0#MORPHO +s.0.ldep#MORPHO +s.0.ldep.ldep#MORPHO +s.0.ldep#LABEL +s.0#DIST.s.1 +s.0#nbl +s.0#LABEL +b.0#LABEL # POS b.-2#POS b.-1#POS -- GitLab