Try out caching some functions to solve spaces issue

app.py

@@ -28,8 +28,24 @@ from transformers import pipeline
 import os
 from transformers_interpret import SequenceClassificationExplainer
 
+
 # USE_model = hub.load("https://tfhub.dev/google/universal-sentence-encoder/4")
-sentence_embedding_model = SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
+
+@st.experimental_singleton
+def get_sentence_embedding_model():
+    return SentenceTransformer('sentence-transformers/all-MiniLM-L6-v2')
+
+@st.experimental_singleton
+def get_spacy():
+    nlp = spacy.load('en_core_web_lg')
+    return nlp
+
+@st.experimental_singleton
+def get_flair_tagger():
+    tagger = SequenceTagger.load("flair/ner-english-ontonotes-fast")
+    return tagger
+
+
 
 # Page setup
 st.set_page_config(
@@ -85,23 +101,16 @@ def display_summary(article_name: str):
     return HTML_WRAPPER.format(soup)
 
 
-##@st.cache(hash_funcs={preshed.maps.PreshMap: my_hash_func})
-def get_spacy():
-    nlp = spacy.load('en_core_web_lg')
-    return nlp
 
 
-# TODO: check the output mutation thingy
-@st.cache(hash_funcs={torch.nn.parameter.Parameter: lambda _: None}, allow_output_mutation=True)
-def get_flair_tagger():
-    tagger = SequenceTagger.load("flair/ner-english-ontonotes-fast")
-    return tagger
+
+
 
 
 def get_all_entities_per_sentence(text):
     # load all NER models
-    nlp = get_spacy()
-    tagger = get_flair_tagger()
+    #nlp = get_spacy()
+    #tagger = get_flair_tagger()
     doc = nlp(text)
 
     sentences = list(doc.sents)
@@ -186,7 +195,7 @@ def render_dependency_parsing(text: str):
 
 # If deps for article: True, otherwise deps for summary calc
 def check_dependency(article: bool):
-    nlp = spacy.load('en_core_web_lg')
+    #nlp = spacy.load('en_core_web_lg')
     if article:
         text = st.session_state.article_text
         all_entities = get_all_entities_per_sentence(text)
@@ -266,6 +275,10 @@ metric, indicating the trustworthiness of the generated summary. Throughout this
 results for some methods on specific examples. These text blocks will be indicated and they change according to the 
 currently selected article.""")
 
+tagger = get_flair_tagger()
+sentence_embedding_model = get_sentence_embedding_model()
+nlp = get_spacy()
+
 # GENERATING SUMMARIES PART
 st.header("Generating summaries")
 st.markdown("Let’s start by selecting an article text for which we want to generate a summary, or you can provide "