QuantumChecker 0.3.2__tar.gz → 0.3.3__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: QuantumChecker
3
- Version: 0.3.2
3
+ Version: 0.3.3
4
4
  Summary: A package to evaluate homework submissions in Python, SQL, PowerBI, and SSIS.
5
5
  Author: Qobiljon
6
6
  Author-email: qobiljonkhayrullayev@gmail.com
@@ -17,14 +17,28 @@ import base64
17
17
 
18
18
  def prompt_text_powerbi(combined_content: str) -> str:
19
19
  return f"""
20
- Evaluate the following Power BI DAX question-answer pairs for correctness, clarity, and appropriateness.
21
- Provide an overall score out of 100 and concise feedback. Focus on DAX logic and structure.
22
- Structure the response as:
23
- OVERALL SCORE: [SCORE]/100
24
- [FEEDBACK]
20
+ You are an expert Power BI instructor evaluating beginner-level DAX question-answer pairs.
21
+
22
+ Each answer contains a data model (in JSON format) extracted from a submitted .pbit file. Evaluate the technical correctness, relevance, and clarity of the DAX elements provided.
23
+
24
+ Use the following criteria to give a fair and supportive evaluation:
25
+ - Measures (40 points): Are calculated measures meaningful, syntactically valid, and aligned with the question?
26
+ - Relationships (20 points): Are key relationships between tables defined logically?
27
+ - Tables & Columns (20 points): Are relevant tables/columns present? Are naming conventions clear?
28
+ - Expressions (10 points): Are query partitions or expressions present and understandable?
29
+ - Overall structure (10 points): Does the model appear coherent and purposeful?
30
+
31
+ **Scoring Tolerance**:
32
+ - Be kind to beginners. Do not give extremely low scores unless the model is completely missing or incorrect.
33
+ - If measures exist and make some sense, award partial credit (e.g., 20–30 out of 40).
34
+ - A score below 30/100 should only be given if there’s little to no relevant content.
35
+
36
+ Structure your response exactly like this:
37
+ OVERALL SCORE: [SCORE]/100
38
+ [Brief feedback here — 3–5 sentences focused on strengths + areas to improve.]
39
+ {combined_content}
40
+ """
25
41
 
26
- {combined_content}
27
- """
28
42
 
29
43
 
30
44
  load_dotenv()
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: QuantumChecker
3
- Version: 0.3.2
3
+ Version: 0.3.3
4
4
  Summary: A package to evaluate homework submissions in Python, SQL, PowerBI, and SSIS.
5
5
  Author: Qobiljon
6
6
  Author-email: qobiljonkhayrullayev@gmail.com
@@ -2,7 +2,7 @@ from setuptools import setup, find_packages
2
2
 
3
3
  setup(
4
4
  name="QuantumChecker",
5
- version="0.3.2",
5
+ version="0.3.3",
6
6
  author="Qobiljon",
7
7
  author_email="qobiljonkhayrullayev@gmail.com",
8
8
  description="A package to evaluate homework submissions in Python, SQL, PowerBI, and SSIS.",
File without changes
File without changes