Update app.py

app.py

@@ -1,30 +1,31 @@
+import streamlit as st
+import numpy as np
+import nltk
+import plotly.express as px
+import plotly.graph_objects as go
+import pandas as pd
 from typing import Dict, List, Union
 from google.cloud import aiplatform
 from google.protobuf import json_format
 from google.protobuf.struct_pb2 import Value
 import os
 import re
-import pandas as pd
-import plotly.express as px
-import plotly.graph_objects as go
-import streamlit as st
-import nltk
 import tempfile
 
-# Process of getting credentials
+# Function to get credentials from environment variable and create a temporary file
 def get_credentials():
-    creds_json_str = os.getenv("JSONSTR")  # Get json credentials stored as a string
+    creds_json_str = os.getenv("JSONSTR")  # Get JSON credentials stored as a string
     if creds_json_str is None:
         raise ValueError("GOOGLE_APPLICATION_CREDENTIALS_JSON not found in environment")
 
     # Create a temporary file
     with tempfile.NamedTemporaryFile(mode="w+", delete=False, suffix=".json") as temp:
-        temp.write(creds_json_str)  # Write in json format
-        temp_filename = temp.name
+        temp.write(creds_json_str)  # Write in JSON format
+        temp_filename = temp.name 
 
     return temp_filename
 
-# Set the credentials
+# Set environment variable for Google application credentials
 os.environ["GOOGLE_APPLICATION_CREDENTIALS"] = get_credentials()
 
 max_seq_length = 2048
@@ -39,6 +40,7 @@ except LookupError:
 
 text_split_tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
 
+# Function to predict emotions using the custom trained model
 def predict_custom_trained_model_sample(
     project: str,
     endpoint_id: str,
@@ -68,7 +70,7 @@ def predict_custom_trained_model_sample(
             split_predictions = clean_prediction.split()
             predictions_list.extend(split_predictions)
         else:
-            print("prediction (unknown type, skipping):", prediction)
+            print(" prediction (unknown type, skipping):", prediction)
     return [emotion for emotion in predictions_list if emotion in d_emotion.values()]
 
 d_emotion = {0: 'admiration', 1: 'amusement', 2: 'anger', 3: 'annoyance', 4: 'approval', 5: 'caring', 6: 'confusion',
@@ -77,28 +79,30 @@ d_emotion = {0: 'admiration', 1: 'amusement', 2: 'anger', 3: 'annoyance', 4: 'ap
              20: 'optimism', 21: 'pride', 22: 'realization', 23: 'relief', 24: 'remorse', 25: 'sadness', 26: 'surprise',
              27: 'neutral'}
 
-st.write("Write or paste any number of document texts to analyze the emotion percentage within your document")
+st.write("Write or paste any number of document texts to analyse the emotion percentage with your document")
 
-# Define user_input outside the conditional block
-user_input = ""
+# Define the sample text
+sample_text = ("Once, in a small village nestled in the rolling hills of Tuscany, lived an elderly woman named Isabella. "
+               "She had spent her entire life in this village, raising her children and caring for her garden, which was the most "
+               "beautiful in the region. Her husband, Marco, had passed away many years ago, leaving her with a heart full of memories "
+               "and a small, quaint house that overlooked the lush vineyards.")
 
 # Add button to fill in sample text
 if st.button("Use Sample Text"):
-    user_input = "Once, in a small village nestled in the rolling hills of Tuscany, lived an elderly woman named Isabella. She had spent her entire life in this village, raising her children and caring for her garden, which was the most beautiful in the region. Her husband, Marco, had passed away many years ago, leaving her with a heart full of memories and a small, quaint house that overlooked the lush vineyards."
+    user_input = sample_text
 else:
     user_input = st.text_area('Enter Text to Analyze')
 
 button = st.button("Analyze")
 
-if user_input and button:
-    alpaca_prompt = """Below is a conversation between a human and an AI agent. Write a response based on the input.
+if button and user_input:
+    alpaca_prompt = """Below is a conversation between a human and an AI agent. write a response based on the input.
         ### Instruction:
         predict the emotion word or words
         ### Input:
         {}
         ### Response:
         """
-
     instances = []
     input_array = text_split_tokenizer.tokenize(user_input)
     for sentence in input_array:
@@ -144,8 +148,6 @@ if user_input and button:
 
     @st.cache_data
     def get_emotion_heatmap(predictions):
-        # Create a matrix for heatmap
-        # Count occurrences of each emotion
         emotion_counts = pd.Series(predictions).value_counts().reset_index()
         emotion_counts.columns = ['Emotion', 'Count']
         
@@ -163,9 +165,9 @@ if user_input and button:
         ))
         fig.update_layout(title='Emotion Heatmap', xaxis_title='Predicted Emotion', yaxis_title='Predicted Emotion')
         return fig
-
+        
     fig_heatmap = get_emotion_heatmap(predictions)
-
+    
     tab1, tab2, tab3 = st.tabs(["Emotion Analysis", "Emotion Counts Distribution", "Heatmap"])
     with tab1:
         st.plotly_chart(fig_pie)