Update app.py

app.py

@@ -3,31 +3,29 @@ import pandas as pd
 import torch
 from transformers import pipeline
 import datetime
-#from datasets import load_dataset
 
-#dataset = load_dataset("wikitablequestions", trust_remote_code=True)
-#item = dataset["test"][10]  # show first test example
+# Load the CSV file and ensure proper formatting
+df = pd.read_csv("anomalies.csv", quotechar='"')
 
-#def to_pandas(item):
-    #return pd.DataFrame(item['table']["rows"], columns=item['table']["header"])
+# Convert 'real' column to standard float format
+df['real'] = df['real'].apply(lambda x: f"{x:.2f}")
 
-#df = to_pandas(item)
-#print(df.head())
-
-df = pd.read_csv("anomalies.csv",quotechar='"',dtype={col: str for col in pd.read_csv('anomalies.csv', nrows=1)})
+# Fill NaN values and convert all columns to strings
 df = df.fillna('').astype(str)
 
-
-# Function to generate a response using the TAPEX model
+# Function to generate a response using the TAPAS model
 def response(user_question, df):
     a = datetime.datetime.now()
 
+    # Initialize the TAPAS model
     tqa = pipeline(task="table-question-answering", model="google/tapas-large-finetuned-wtq")
 
+    # Debugging information
     print("DataFrame shape:", df.shape)
     print("DataFrame head:\n", df.head())
     print("User question:", user_question)
 
+    # Query the TAPAS model
     answer = tqa(table=df, query=user_question)['answer']
 
     query_result = {
@@ -35,7 +33,7 @@ def response(user_question, df):
     }
 
     b = datetime.datetime.now()
-    print(b - a)
+    print("Time taken:", b - a)
 
     return query_result
 
@@ -77,4 +75,4 @@ for sender, message in st.session_state['history']:
     if sender == '👤':
         st.markdown(f"**👤 {message}**")
     elif sender == '🤖':
-        st.markdown(f"<div style='text-align: right'>**🤖 {message}**</div>", unsafe_allow_html=True)
+        st.markdown(f"<div style='text-align: right'>**🤖 {message}**</div>", unsafe_allow_html=True)