From 6e42d70f1844150f0327a436bca856e7a527de01 Mon Sep 17 00:00:00 2001
From: Jacques Fize <jacques.fize@insa-lyon.fr>
Date: Tue, 3 Nov 2020 11:44:25 +0100
Subject: [PATCH] UPD

---
 README.md                                        |  3 +--
 generate_dataset.py                              |  1 +
 .../toponym_combination_embedding_v3.json        |  3 ++-
 train_geocoder_v2.py                             | 16 ++++++++++++----
 4 files changed, 16 insertions(+), 7 deletions(-)

diff --git a/README.md b/README.md
index f402d35..f5c5288 100644
--- a/README.md
+++ b/README.md
@@ -187,8 +187,7 @@ grid = GridSearchModel(\
 grid.run()
 ```
 
-
-# Authors and Acknowledgment
+# Authors and Acknowledgment
 
 Proposed by **Jacques Fize**, **Ludovic Moncla** and **Bruno Martins**
 
diff --git a/generate_dataset.py b/generate_dataset.py
index 75713f0..788fe37 100644
--- a/generate_dataset.py
+++ b/generate_dataset.py
@@ -23,6 +23,7 @@ parser.add_argument("--split-method", default="per_pair", type=str, choices="per
 args = parser.parse_args()#("../data/geonamesData/FR.txt ../data/wikipedia/cooccurrence_FR.txt ../data/geonamesData/hierarchy.txt".split())
 
 PREFIX = args.geonames_dataset.split("/")[-1].split(".")[0]  # Ouch !
+PREFIX = PREFIX + "_" + args.split_method
 
 #  LOAD DATA
 geonames_data = read_geonames(args.geonames_dataset)
diff --git a/parser_config/toponym_combination_embedding_v3.json b/parser_config/toponym_combination_embedding_v3.json
index 5053a89..fac6e68 100644
--- a/parser_config/toponym_combination_embedding_v3.json
+++ b/parser_config/toponym_combination_embedding_v3.json
@@ -13,6 +13,7 @@
         { "long": "--ngram-word2vec-iter", "type": "int", "default": 50 },
         { "short": "-t", "long": "--tolerance-value", "type": "float", "default": 100 },
         { "short": "-e", "long": "--epochs", "type": "int", "default": 100 },
-        { "short": "-d", "long": "--dimension", "type": "int", "default": 256 }
+        { "short": "-d", "long": "--dimension", "type": "int", "default": 256 },
+        { "short": "-l", "long": "--lstm-layer", "type": "int", "default": 2,"choices":[1,2] }
     ]
 }
\ No newline at end of file
diff --git a/train_geocoder_v2.py b/train_geocoder_v2.py
index e5138f8..e67f311 100644
--- a/train_geocoder_v2.py
+++ b/train_geocoder_v2.py
@@ -62,7 +62,7 @@ if args.adjacency:
     PREFIX_OUTPUT_FN += "_A"
 if args.inclusion:
     PREFIX_OUTPUT_FN += "_I"
-if args.wikipedia_cooc:
+if args.wikipedia:
     PREFIX_OUTPUT_FN += "_C"
 
 MODEL_OUTPUT_FN = "outputs/{0}.h5".format(PREFIX_OUTPUT_FN)
@@ -170,9 +170,15 @@ x1 = embedding_layer(input_1)
 x2 = embedding_layer(input_2)
 
 # Each LSTM learn on a permutation of the input toponyms
-x1 = Bidirectional(LSTM(100))(x1)
-x2 = Bidirectional(LSTM(100))(x2)
-x = concatenate([x1,x2])
+if args.lstm_layer == 2:
+    x1 = Bidirectional(LSTM(100))(x1)
+    x2 = Bidirectional(LSTM(100))(x2)
+    x = concatenate([x1,x2])
+else:
+    lstm_unique_layer = Bidirectional(LSTM(100))
+    x1 = lstm_unique_layer(x1)
+    x2 = lstm_unique_layer(x2)
+    x = concatenate([x1,x2])
 
 x1 = Dense(500,activation="relu")(x)
 x1 = Dense(500,activation="relu")(x1)
@@ -188,6 +194,8 @@ output_coord = concatenate([output_lon,output_lat],name="output_coord")
 model = Model(inputs = [input_1,input_2], outputs = output_coord)#input_3
 model.compile(loss={"output_coord":haversine_tf_1circle}, optimizer='adam',metrics={"output_coord":accuracy_k(ACCURACY_TOLERANCE)})
 
+print("Neural Network Architecture : ")
+print(model.summary())
 #############################################################################################
 ################################# TRAINING LAUNCH ###########################################
 #############################################################################################
-- 
GitLab