Update app.py

app.py

@@ -29,7 +29,7 @@ def read_pdf(file):
   # We use BM25 as retriver which will do 1st round of candidate filtering based on word based matching
 
 def bm25_tokenizer(text):
-    stop_w = ['a', 'the', 'am', 'is' , 'are', 'who', 'how', 'where', 'when', 'why']
+    stop_w = ['a', 'the', 'am', 'is' , 'are', 'who', 'how', 'where', 'when', 'why', 'what']
     tokenized_doc = []
     for token in text.lower().split():
         token = token.strip(string.punctuation)
@@ -84,8 +84,8 @@ def final_qa_pipeline(file, query):
         
 examples = [
     [os.path.abspath("NASDAQ_AAPL_2020.pdf"), "how much are the outstanding shares?"],
-    [os.path.abspath("NASDAQ_MSFT_2020.pdf"), "how much was the profit?"],
-    [os.path.abspath("dbs-annual-report-2020.pdf"), "What is the shareholder equity?"],
+    [os.path.abspath("NASDAQ_MSFT_2020.pdf"), "what was the income before taxes?"],
+    [os.path.abspath("dbs-annual-report-2020.pdf"), "how much dividend was paid to shareholders??"],
 ]
 
 iface = gr.Interface(