owusu/erp-ai-latest
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

updated questions generator

Browse Source
This commit is contained in:
kowshik
2024-10-15 01:56:05 +00:00
parent cd10cdaf7d
commit 8ce331b023
10 changed files with 427 additions and 105 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/prompts/questions.py
+73 -26
View File
@@ -65,39 +65,86 @@ def get_questions_prompt_v2():
prompt = """
You are tasked with generating assessment questions for workers based on their SOPs. These questions should vary by assessment type (daily, weekly, biweekly), frequency number (e.g., day 2, week 3), and total duration (e.g., 6 weeks).
You are generating a variable number of assessment questions for different assessment types, which can be daily, weekly, biweekly, etc. The assessment frequency and the specific frequency number should be taken into account when generating the questions.
For a particular assessment, based on the type (e.g., daily, weekly, biweekly) and total duration (e.g., if it is weekly and the total duration is 6 weeks, generate at least 20 questions weekly for up to six weeks).
Each question will be based on the SOPs of specific workers in different departments, and the questions should vary depending on the assessment frequency type and frequency number.
The goal of the assessment is to focus on the progress of the tasks outlined in each worker's SOPs.
Make sure each question is relevant to the worker's SOP, and attach an area tag ID, aissgned_to ID, role ID, beacuse each role sop will the provided with area tags and member for that role, so questions should be generated based on that area
The questions should become more detailed or challenging as the assessment progresses over time and the question is
Guidelines:
1. Generate yes/no questions relevant to each worker's SOP based on their role.
2. Use frequency numbers as 1, 2, 3, etc., regardless of the type.
3. Tag each question with its topic area (e.g., communication, timeline).
3. Tag each question with the provided topic area ID (e.g., 1, 2, 3, 4, 5).
4. Tag each question with the provided role ID (e.g., 1, 2, 3, 4, 5).
5. Tag each question with the provided assigned_to ID (the ID of the person in charge) (e.g., 1, 2, 3, 4, 5).
4. The questions should evolve in detail as assessments progress over time.
6. For example if ans assement_type is weekly and the duration is 4, this means we generate questions for 4 weeks , frequency_number 1 to top 4 using the sop and all data
7 For each frequency_number, generate at least 15 - 20 questions.
Example response:
{
{
"questions": {
"questions": [
{
"frequency_number": "2",
"questions": [
{
"assigned_to": "John",
"role": "Data Analyst",
"question": "Is the dataset being cleaned according to the SOP?",
"area_tag": "data quality"
}
]
},
{
"frequency_number": "3",
"questions": [
{
"assigned_to": "Jane",
"role": "Data Scientist",
"question": "Are predictive models tested for accuracy before deployment?",
"area_tag": "model validation"
}
]
}
]
}
{
frequency_number: 1,
items: [
{
"area_tag": 5,
"assigned_to": 8,
"questions": "Has the content calendar been developed and shared with the team?",
"role": 4
},
{
}
]
},
{
     }
        ]
"""
return prompt
def get_questions_prompt_v3():
prompt = """
You are tasked with generating assessment questions for workers based on their SOPs. These questions should vary by assessment type (daily, weekly, biweekly), frequency number (e.g., day 2, week 3), and total duration (e.g., 6 weeks).
Guidelines:
1. Generate yes/no questions relevant to each worker's SOP based on their role.
2. Use frequency numbers as 1, 2, 3, etc., regardless of the type.
3. Tag each question with the provided topic area ID (e.g., 1, 2, 3, 4, 5).
4. Tag each question with the provided role ID (e.g., 1, 2, 3, 4, 5).
5. Tag each question with the provided assigned_to ID (the ID of the person in charge) (e.g., 1, 2, 3, 4, 5).
6. The questions should evolve in detail as assessments progress over time.
7. For each frequency_number, generate at least 15 - 20 questions.
Provide the response in the following JSON format:
{
"questions": {
"questions": [
{
"frequency_number": 1,
"items": [
{
"area_tag": 5,
"assigned_to": 8,
"questions": "Has the content calendar been developed and shared with the team?",
"role": 4
},
...
]
},
...
]
}
}
"""
return prompt