From 35d2af6a15ec2070dd3c28e5d28212c42770b7fe Mon Sep 17 00:00:00 2001
From: Yigit Sever
Date: Fri, 27 Sep 2019 22:04:33 +0300
Subject: Include batch scripts

---
 README.md         | 17 ++++++++++++++---
 run_semb.sh       | 18 ++++++++++++++++++
 run_supervised.sh | 16 ++++++++++++++++
 run_wmd.sh        | 18 ++++++++++++++++++
 4 files changed, 66 insertions(+), 3 deletions(-)
 create mode 100755 run_semb.sh
 create mode 100755 run_supervised.sh
 create mode 100755 run_wmd.sh

diff --git a/README.md b/README.md
index e80d946..ff2256d 100644
--- a/README.md
+++ b/README.md
@@ -157,7 +157,10 @@ Example;
 python WMD.py en bg bilingual_embeddings/en_to_bg.vec bilingual_embeddings/bg_to_en.vec wordnets/ready/en_to_bg.def wordnets/ready/bg_to_en.def wmd retrieval
 ```
 
-Will run on English and Bulgarian definitions, using WMD for retrieval.
+Will run on English and Bulgarian definitions, using WMD for retrieval. We included a batch script to run WMD and SNK with retrieval and matching on all available language pairs;
+```
+./run_wmd.sh
+```
 
 ### sentence_embedding.py - Sentence Embedding Representation
 
@@ -193,7 +196,11 @@ Example;
 python sentence_embedding.py it ro bilingual_embeddings/it_to_ro.vec bilingual_embeddings/ro_to_it.vec wordnets/ready/it_to_ro.def wordnets/ready/ro_to_it.def matching
 ```
 
-Will run on Italian and Romanian definitions, using sentence embedding representation for matching.
+Will run on Italian and Romanian definitions, using sentence embedding representation for matching. We included a batch script to run alignment using sentence embeddings using retrieval and matching on all available language pairs;
+
+```
+./run_semb.sh
+```
 
 
 ### learn_and_predict.py - Supervised Alignment
@@ -234,4 +241,8 @@ Example;
 python learn_and_predict.py -sl en -tl ro -df ./wordnets/tsv_files/en_to_ro.tsv -es bilingual_embeddings/en_to_ro.vec -et bilingual_embeddings/ro_to_en.vec
 ```
 
-Will run on English and Romanian definitions.
+Will run on English and Romanian definitions. We included a batch script to run supervised alignment on all available pairs;
+
+```
+./run_supervised.sh
+```
diff --git a/run_semb.sh b/run_semb.sh
new file mode 100755
index 0000000..7256fbf
--- /dev/null
+++ b/run_semb.sh
@@ -0,0 +1,18 @@
+#!/bin/bash
+
+set -o errexit -o pipefail -o noclobber -o nounset
+
+ROOTDIR="$(pwd)"
+ready_vectors_path="${ROOTDIR}/bilingual_embeddings"
+wordnets_path="${ROOTDIR}/wordnets/ready"
+
+for i in  en,bg en,el en,it, en,ro, en,sl en,sq, bg,el bg,it bg,ro el,it el,ro el,sq it,ro ro,sl ro,sq; do
+    IFS=',' read -r source_lang target_lang <<< "${i}"
+    echo "SEMB: ${source_lang} - ${target_lang}"
+    source_vec="${ready_vectors_path}/${source_lang}_to_${target_lang}.vec"
+    target_vec="${ready_vectors_path}/${target_lang}_to_${source_lang}.vec"
+    source_def="${wordnets_path}/${source_lang}_to_${target_lang}.def"
+    target_def="${wordnets_path}/${target_lang}_to_${source_lang}.def"
+    python "${ROOTDIR}/sentence_embedding.py" "${source_lang}" "${target_lang}" "${source_vec}" "${target_vec}" "${source_def}" "${target_def}" all -n 1000 -b
+    sleep 5
+done
diff --git a/run_supervised.sh b/run_supervised.sh
new file mode 100755
index 0000000..30cc548
--- /dev/null
+++ b/run_supervised.sh
@@ -0,0 +1,16 @@
+#!/bin/bash
+
+set -o errexit -o pipefail -o noclobber -o nounset
+
+ROOTDIR="$(pwd)"
+ready_vectors_path="${ROOTDIR}/bilingual_embeddings"
+tsv_path="${ROOTDIR}/wordnets/tsv_files"
+
+for i in  en,bg en,el en,it, en,ro, en,sl en,sq, bg,el bg,it bg,ro el,it el,ro el,sq it,ro ro,sl ro,sq; do
+    IFS=',' read -r source_lang target_lang <<< "${i}"
+    source_vec="${ready_vectors_path}/${source_lang}_to_${target_lang}.vec"
+    target_vec="${ready_vectors_path}/${target_lang}_to_${source_lang}.vec"
+    data_file="${tsv_path}/${source_lang}_to_${target_lang}.tsv"
+    python "${ROOTDIR}/learn_and_predict.py" -sl "${source_lang}" -tl "${target_lang}" -df "${data_file}" -es "${source_vec}" -et "${target_vec}" -b
+    sleep 5
+done
diff --git a/run_wmd.sh b/run_wmd.sh
new file mode 100755
index 0000000..cb79f70
--- /dev/null
+++ b/run_wmd.sh
@@ -0,0 +1,18 @@
+#!/bin/bash
+
+set -o errexit -o pipefail -o noclobber -o nounset
+
+ROOTDIR="$(pwd)"
+ready_vectors_path="${ROOTDIR}/bilingual_embeddings"
+wordnets_path="${ROOTDIR}/wordnets/ready"
+
+for i in  en,bg en,el en,it, en,ro, en,sl en,sq, bg,el bg,it bg,ro el,it el,ro el,sq it,ro ro,sl ro,sq; do
+    IFS=',' read -r source_lang target_lang <<< "${i}"
+    echo "WMD + SNK: ${source_lang} - ${target_lang}"
+    source_vec="${ready_vectors_path}/${source_lang}_to_${target_lang}.vec"
+    target_vec="${ready_vectors_path}/${target_lang}_to_${source_lang}.vec"
+    source_def="${wordnets_path}/${source_lang}_to_${target_lang}.def"
+    target_def="${wordnets_path}/${target_lang}_to_${source_lang}.def"
+    python "${ROOTDIR}/WMD.py" "${source_lang}" "${target_lang}" "${source_vec}" "${target_vec}" "${source_def}" "${target_def}" all all -n 1000 -b
+    sleep 5
+done
-- 
cgit v1.2.3-61-g4310