cache_resource using streamlit

app.py

@@ -4,39 +4,41 @@ import pandas
 from sentence_transformers import SentenceTransformer, util
 import torch
 
-# from sentence_transformers import SentenceTransformer, util
-
-
 st.title('Arxiv Paper Recommendation')
 paper_you_like = st.text_input(
         "Enter the title of any paper you like 👇",
-    placeholder  = "Attention is all you need"
+    placeholder  = None
     )
 
-# print(paper_you_like)
+@st.cache_resource
+def get_sentences_data():
+    with open('sentences.pkl', 'rb') as f:
+        sentences = pickle.load(f)
+    return sentences
 
-with open('sentences.pkl', 'rb') as f:
-    sentences = pickle.load(f)
+sentences = get_sentences_data()
 
-with open('embeddings.pkl', 'rb') as f:
-    embeddings = pickle.load(f)
 
+@st.cache_resource
+def get_embeddings_data():
+    with open('embeddings.pkl', 'rb') as f:
+        embeddings = pickle.load(f)
+    return embeddings
 
-model = SentenceTransformer('all-MiniLM-L6-v2')
+embeddings = get_embeddings_data()
 
-# Calculating the similarity between titles
-cosine_scores = util.cos_sim(embeddings, model.encode(paper_you_like))
 
-# print(cosine_scores)
+@st.cache_resource
+def get_model():
+    model = SentenceTransformer('all-MiniLM-L6-v2')
+    return model
 
+model = get_model()
 
+# Calculating the similarity between titles
+cosine_scores = util.cos_sim(embeddings, model.encode(paper_you_like))
 top_similar_papers = torch.topk(cosine_scores,dim=0, k=5,sorted=True)
 # top_similar_papers
 
-# s = ''
-
 for i in top_similar_papers.indices:
-#     print(i)
-    st.write(sentences[i.item()])
-# st.text(s)
-
+    st.write(sentences[i.item()])