This repository serves as a replication package for the research paper titled "Typestate-based Fault Localization of API Usage Violations in a Deep Learning Program"

Overview This repository serves as a replication package for the research paper titled "An Approach to Detecting Usage Protocol Violations in Deep Learning Programs." Benchmarks Two benchmarks were utilized in the research: Benchmark1 from NeuraLint: Location: NLBench/SOSamples Benchmark2 from Humbatova et al.: Location: HumbatovaBench/SOSamples Reproducing Results To replicate the results presented in the paper, follow these steps: Download the NeuralStateAnalysis Zip file. Extract the file and navigate to the NeuralStateAnlaysis directory. (Optional) Install the necessary requirements by executing pip install requirements.txt. Note: The requirements.txt file is already available in this repository. Running NeuralState on NLBench: Navigate to the NLBench/SOSamples directory. Open any of the programs you wish to execute. Set the path to the NeuralStateAnalysis directory: Path-to-folder/NeuralStateAnalysis/ Run the program using the command python program_id. Since the 'NeuralStateAnalysis(model).debug()' call is already present in all programs, you will be able to reproduce the results. Running NeuralState on HumbatovaBench: Navigate to the HumbatovaBench/SOSamples directory. Open any of the programs you wish to execute. Set the path to the NeuralStateAnalysis directory: Path-to-folder/NeuralStateAnalysis/ Run the program using the command python program_id. Since the 'NeuralStateAnalysis(model).debug()' call is already present in all programs, you will be able to reproduce the results.