updated to singletons

.vscode/launch.json

@@ -1,16 +0,0 @@
-{
-    // Use IntelliSense to learn about possible attributes.
-    // Hover to view descriptions of existing attributes.
-    // For more information, visit: https://go.microsoft.com/fwlink/?linkid=830387
-    "version": "0.2.0",
-    "configurations": [
-        {
-            "name": "Python: Current File",
-            "type": "python",
-            "request": "launch",
-            "program": "${file}",
-            "console": "integratedTerminal",
-            "justMyCode": false
-        }
-    ]
-}

.vscode/settings.json

@@ -1,7 +0,0 @@
-{
-    "workbench.colorCustomizations": {
-        "activityBar.background": "#09323E",
-        "titleBar.activeBackground": "#0C4656",
-        "titleBar.activeForeground": "#F6FCFE"
-    }
-}

rebel.py

@@ -27,8 +27,23 @@ DEFAULT_LABEL_COLORS = {
     "PERCENT": "#e4e7d2",
 }
 
-def generate_knowledge_graph(texts: List[str], filename: str):
+
+@st.experimental_singleton(max_entries=1)
+def get_pipeline():
+    triplet_extractor = pipeline('text2text-generation', model='Babelscape/rebel-large', tokenizer='Babelscape/rebel-large')
+    return triplet_extractor
+
+
+
+@st.experimental_singleton(max_entries=1)
+def load_spacy():
     nlp = spacy.load("en_core_web_sm")
+    return nlp
+
+
+def generate_knowledge_graph(texts: List[str], filename: str):
+    nlp = load_spacy()
+
     doc = nlp("\n".join(texts).lower())
     NERs = [ent.text for ent in doc.ents]
     NER_types =  [ent.label_ for ent in doc.ents]
@@ -78,7 +93,7 @@ def generate_knowledge_graph(texts: List[str], filename: str):
 
 @lru_cache(maxsize=16)
 def generate_partial_graph(text: str):
-    triplet_extractor = pipeline('text2text-generation', model='Babelscape/rebel-large', tokenizer='Babelscape/rebel-large')
+    triplet_extractor = get_pipeline()
     a = triplet_extractor(text, return_tensors=True, return_text=False)[0]["generated_token_ids"]["output_ids"]
     extracted_text = triplet_extractor.tokenizer.batch_decode(a)
     extracted_triplets = extract_triplets(extracted_text[0])

utils.py

@@ -3,4 +3,6 @@ def clip_text(t, lenght = 4):
     t_sub = t.replace("...", "dotdotdot")
     t_clipped = ".".join(t_sub.split(".")[:lenght]) + "."
     t_reverted = t_clipped.replace("dotdotdot", "...")
-    return t_reverted
+    return t_reverted
+
+