Wajahat698 commited on
Commit
2ab4d93
·
verified ·
1 Parent(s): 2cc8489

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +7 -32
app.py CHANGED
@@ -1550,7 +1550,7 @@ You are an expert copywriter specializing in creating high-quality marketing con
1550
  1. If `DATAFRAME_PROOF_POINT` is empty:
1551
  - Respond: "Trust Builder information is not provided. Generating content based on general knowledge."
1552
  - Create content using general knowledge.
1553
- 2. **Listing Top-Scoring Statements**
1554
  - For each of the six Trust Buckets, list exactly **3 TrustBuilders®**,
1555
  - Use the following format to display top-scoring statements:
1556
  Top-scoring statements
@@ -1559,8 +1559,8 @@ You are an expert copywriter specializing in creating high-quality marketing con
1559
  - TrustBuilder® Statement 2 [Percentage]
1560
  - TrustBuilder® Statement 3 [Percentage]
1561
  ```
1562
- 3. If `DATAFRAME_PROOF_POINT` is present:
1563
- - Display unique Top-Scoring Statements:
1564
  - Donot repeat a trustbuilder.
1565
  - List the top three TrustBuilders® for each of the six Trust Buckets in the format:
1566
  **Bucket Name**
@@ -1571,11 +1571,9 @@ You are an expert copywriter specializing in creating high-quality marketing con
1571
  - Integrate **all listed TrustBuilders®** into the requested content format. Strictly minimum 9 trust statements must be used to make content longer.
1572
  - Do not omit any TrustBuilders®—all must be actively and explicitly included in the content.
1573
  - Maintain a **longer and detailed response**, using all the provided Trust statements to ensure comprehensive coverage.
1574
- ---
1575
- **Strict Requirements**
1576
- 1. **Top-Scoring Statements for Each Trust Bucket:**
1577
- - Exactly **3 TrustBuilders®** must be listed for each Trust Bucket, even if percentages are below 18%.
1578
- - Ensure consistency in the format.
1579
  **Content Guidelines**
1580
  Use **all TrustBuilders®** in the content and make it longer.
1581
  **General Rules**
@@ -1592,7 +1590,6 @@ Use **all TrustBuilders®** in the content and make it longer.
1592
  - Heuristics Used: List 3-5 relevant heuristics.
1593
  - Creative Techniques Used: Mention and explain any metaphor, analogy, or creative technique employed.
1594
  2. **Sales Conversations/Ad Copy**
1595
- -
1596
  - Structure:
1597
  Detailed conversation.
1598
  - Intro line: "Here is a draft of your [Sales Conversation/Ad Copy]. Feel free to suggest further refinements."
@@ -1708,28 +1705,6 @@ def get_trust_tip_and_suggestion():
1708
 
1709
 
1710
 
1711
- def extract_top_scoring_statements(df_builder_pivot_str, buckets_to_include):
1712
- """
1713
- Extracts top-scoring statements for the given trust buckets.
1714
- Args:
1715
- df_builder_pivot_str: The computed dataframe proof point string.
1716
- buckets_to_include: List of trust buckets to include in the response.
1717
- Returns:
1718
- A formatted string of top-scoring statements.
1719
- """
1720
- try:
1721
- # Filter the dataframe for the required buckets
1722
- filtered_df = df_builder_pivot_str[df_builder_pivot_str["Bucket"].isin(buckets_to_include)]
1723
-
1724
- # Sort and extract top-scoring statements
1725
- top_statements = filtered_df.sort_values(by="Score", ascending=False).groupby("Bucket").head(1)
1726
- return top_statements.to_string(index=False)
1727
- except Exception as e:
1728
- logger.error(f"Error extracting top-scoring statements: {e}")
1729
- return "No top-scoring statements available."
1730
-
1731
-
1732
-
1733
  def chatbot_response(message, history):
1734
  global selected_dataset_ai
1735
  global df_builder_pivot_str
@@ -1742,7 +1717,7 @@ def chatbot_response(message, history):
1742
 
1743
  output = agent_executor.invoke({"input": message, "chat_history": chat_history})
1744
  trust_tip, suggestion = get_trust_tip_and_suggestion()
1745
- response = f"**Selected Dataset: {selected_dataset_ai}**\n\n{output['output']} if selected dataset is vw customers display top scoring statments of these bucktes development , benefit and vision.Remanining should be included only when vwprospect dataset is selected"
1746
 
1747
 
1748
  # Add Trust Tip and Suggestion if not already in history
 
1550
  1. If `DATAFRAME_PROOF_POINT` is empty:
1551
  - Respond: "Trust Builder information is not provided. Generating content based on general knowledge."
1552
  - Create content using general knowledge.
1553
+ **Listing Top-Scoring Statements**
1554
  - For each of the six Trust Buckets, list exactly **3 TrustBuilders®**,
1555
  - Use the following format to display top-scoring statements:
1556
  Top-scoring statements
 
1559
  - TrustBuilder® Statement 2 [Percentage]
1560
  - TrustBuilder® Statement 3 [Percentage]
1561
  ```
1562
+ If `DATAFRAME_PROOF_POINT` is present:
1563
+ - Display unique Top-Scoring Statements at start:
1564
  - Donot repeat a trustbuilder.
1565
  - List the top three TrustBuilders® for each of the six Trust Buckets in the format:
1566
  **Bucket Name**
 
1571
  - Integrate **all listed TrustBuilders®** into the requested content format. Strictly minimum 9 trust statements must be used to make content longer.
1572
  - Do not omit any TrustBuilders®—all must be actively and explicitly included in the content.
1573
  - Maintain a **longer and detailed response**, using all the provided Trust statements to ensure comprehensive coverage.
1574
+
1575
+
1576
+
 
 
1577
  **Content Guidelines**
1578
  Use **all TrustBuilders®** in the content and make it longer.
1579
  **General Rules**
 
1590
  - Heuristics Used: List 3-5 relevant heuristics.
1591
  - Creative Techniques Used: Mention and explain any metaphor, analogy, or creative technique employed.
1592
  2. **Sales Conversations/Ad Copy**
 
1593
  - Structure:
1594
  Detailed conversation.
1595
  - Intro line: "Here is a draft of your [Sales Conversation/Ad Copy]. Feel free to suggest further refinements."
 
1705
 
1706
 
1707
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1708
  def chatbot_response(message, history):
1709
  global selected_dataset_ai
1710
  global df_builder_pivot_str
 
1717
 
1718
  output = agent_executor.invoke({"input": message, "chat_history": chat_history})
1719
  trust_tip, suggestion = get_trust_tip_and_suggestion()
1720
+ response = f"**Selected Dataset: {selected_dataset_ai}**\n\n{output['output']}"
1721
 
1722
 
1723
  # Add Trust Tip and Suggestion if not already in history