DEBUG

combination_embeddings.py

@@ -50,7 +50,7 @@ logging.basicConfig(
     )
 chrono = Chronometer()
 
-args = ConfigurationReader("./parser_config/toponym_combination_embedding.json").parse_args("-i -a -n 2 -t 0.002 -e 5 -m CNN data/geonamesData/FR.txt data/geonamesData/hierarchy.txt".split())
+args = ConfigurationReader("./parser_config/toponym_combination_embedding.json").parse_args()#("-i -a -n 2 -t 0.002 -e 5 -m CNN data/geonamesData/FR.txt data/geonamesData/hierarchy.txt".split())
 
 GEONAME_FN = args.geoname_input
 GEONAMES_HIERARCHY_FN = args.geoname_hierachy_input
@@ -85,7 +85,8 @@ bounds = filtered.dissolve("i").bounds.values[0]
 rel_dict ={}
 
 if args.adjacency:
-    fn = "{0}_adjacency.json".format(GEONAME_FN.split("/")[-1])
+    logging.info("Retrieve inclusion relationships ! ")
+    fn = "data/geonamesData/{0}_adjacency.json".format(GEONAME_FN.split("/")[-1])
     if not os.path.exists(fn):
         g = Grid(*bounds,[360,180])
         g.fit_data(filtered)
@@ -93,7 +94,9 @@ if args.adjacency:
         rel_dict.update(dict([[int(i) for i in r.split("|")] for r in g.get_adjacent_relationships()]))
         json.dump(rel_dict,open(fn,'w'))
     else:
-        rel_dict.update(json.load(open(fn,'w')))
+        logging.info("Open and load data from previous computation!")
+        rel_dict.update({int(k):int(v) for k,v in json.load(open(fn)).items()})
+    logging.info("{0} adjacency relationships retrieved ! ".format(len(rel_dict)))
 
 if args.inclusion:
     # RETRIEVE INCLUSION RELATIONSHIPS
@@ -176,6 +179,11 @@ def accuracy_at_k(y_true, y_pred):
     return K.size(fit[:,0])/K.size(y_pred),K.size(fit[:,1])/K.size(y_pred)
 
 name = "{0}_{1}_{2}_{3}".format(GEONAME_FN.split("/")[-1],EPOCHS,NGRAM_SIZE,ACCURACY_TOLERANCE)
+if args.adjacency:
+    name+="_A"
+if args.inclusion:
+    name+="_I"
+
 logging.info("Generating N-GRAM Embedding...")
 embedding_weights = index.get_embedding_layer(geoname2encodedname.values(),dim= embedding_dim)
 logging.info("Embedding generated !")
@@ -212,7 +220,7 @@ if CONV :
     input_2 = Input(shape=(max_len,))
     #input_3 = Input(shape=(1,))
 
-    embedding_layer = Embedding(num_words, embedding_dim,input_length=max_len, trainable=True)
+    embedding_layer = Embedding(num_words, embedding_dim,input_length=max_len, weights=[embedding_weights],trainable=False)
 
     x1 = Conv1D(filters=32, kernel_size=3, activation='relu')(embedding_layer(input_1))
     x1 = Dropout(0.5)(x1)

evalgeonamesembeddings.py

@@ -43,7 +43,7 @@ df["geometry"] = df["latitude longitude".split()].apply(lambda x:Point(x.longitu
 gdf = gpd.GeoDataFrame(df)
 
 # Select a sample that concerns the departement "La Manche"
-manche_gdf = gdf[gdf.admin2_code == "50"]
+manche_gdf = gdf[gdf.admin2_code == "50"].copy()
 
 df =pd.DataFrame([get_data(fn) for fn in fns])
 
@@ -61,10 +61,10 @@ def get_pearsons(model):
     return pearsonr(geodesic_d , embeddings_d) # Compute Pearson correlation and associated p-value
 
 df["pearson"] = df.filepath.apply(lambda x : get_pearsons(gensim.models.KeyedVectors.load(x))[0])
-
-df.plot.scatter(x="walk_length", y="pearson")
+df.fillna(0,inplace=True)
+df.plot.scatter(x="walk_length", y="pearson",c="noise",cmap='inferno')
 plt.show()
-df.plot.scatter(x="number_of_walks", y="pearson")
+df.plot.scatter(x="number_of_walks", y="pearson",c="noise",cmap='inferno')
 plt.show()
-df.plot.scatter(x="word2vec_window_size", y="pearson")
+df.plot.scatter(x="word2vec_window_size", y="pearson",c="noise",cmap='inferno')
 plt.show()
\ No newline at end of file