CognitioMachina aims to categorize a program's activity using a deep learning model without significantly impacting the program itself. The training process involves initiating the profiled program and executing the targeted functions or activities of interest. Subsequently, the runtimes of the pertinent software components are logged and stored in .csv files. This data is then utilized in a training session to generate a PyTorch model. In the subsequent runtime profiling, a Linux pipe can be set up between the profiled program and the deep learning model, facilitating the prediction of program activity based on the real-time function runtimes.
Features
- AI
- Deep Learning
- C++
- PyTorch
- Python
- Profiler
- Linux
- LSTM
Follow CognitioMachina
You Might Also Like
Rate This Project
Login To Rate This Project
User Reviews
Be the first to post a review of CognitioMachina!