Updated to read files from the web

app.py

@@ -8,6 +8,9 @@ from collections import OrderedDict
 from sentence_transformers import SentenceTransformer, CrossEncoder, util
 import torch
 
+import requests
+import io
+
 from nltk.tokenize import sent_tokenize
 
 import nltk
@@ -47,8 +50,9 @@ def sent_cross_load():
 
 @st.cache
 def load_data():
-    dfs = [pd.read_json(f'passages_{i}.jsonl', lines=True) for i in range(0,5)]
-    df = pd.concat(dfs)
+    #dfs = [pd.read_json(f'passages_{i}.jsonl', lines=True) for i in range(0,5)]
+    #df = pd.concat(dfs)
+    df = pd.read_json('https://www.dropbox.com/s/82lwbaym3b1o6uq/passages.jsonl?raw=1', lines=True)
     df.reset_index(inplace=True, drop=True)
     return df
 
@@ -59,8 +63,12 @@ with st.spinner(text="Loading data..."):
 
 @st.cache
 def load_embeddings():
-    efs = [np.load(f'embeddings_{i}.pt.npy')  for i in range(0,5)]
-    corpus_embeddings = np.concatenate(efs)
+    #efs = [np.load(f'embeddings_{i}.pt.npy')  for i in range(0,5)]
+    #corpus_embeddings = np.concatenate(efs)
+    response = requests.get('"https://www.dropbox.com/s/px8kjdd3p5mzw6j/corpus_embeddings.pt.npy?raw=1"')
+    corpus_embeddings = np.load(io.BytesIO(response.content))  
+
+    
     return corpus_embeddings
 
 with st.spinner(text="Loading embeddings..."):