The PredictorProjGPT prompt framework has the capability to accept performance test results as an input and provide an output that includes Key Performance Indicators (KPIs) mapped to their corresponding dataframe objects.
OpenAI Documentation : https://platform.openai.com/docs/libraries/python-library
1. OpenAI API Key
2. OpenAI Packages
3. Pandas Packages
4. Streamlit Packages
1. Install OpenAI Packages
pip install openai
2. Install Pandas
pip install pandas
3. Install Streamlit
pip install streamlit
A performance test was conducted on 24th April 2020 on a Salesforce application. The following observation was
recorded by a performance tester after conclusion of the test.
Configuration:
1. There are total 40 transactions involved. The wait time between each transaction is 5 seconds.
2. Total Pacing Time is configured as 180 seconds.
Summary:
1. Average Response Time for the entire duration of the test was 6.5 seconds.
2. CPU Utilization peaked 75% in the first hour of the steady state.
3. Memory Consumption of 88% was observed at the last hour of the steady state.
4. Entire duration of the test was 5 hours. Endurance Test was performed.
5. Observed 10% error on the entire duration of the test
6. Predict the outcome
1. Clone the repository into your project directory
git clone 'enter repository url here'
2. Build Image
docker build -t predictor-gpt-app .
2. Run the container
docker run -it predictor-gpt-app
1. Recent Changes in OpenAI commands - https://github.com/openai/openai-python
2. OpenAI secret key is now stored as environment variable