BinAdapter prototype

For base model implementation please refer here

Steps for training AsmDepictor with incremental data:

Install prerequisites:

Working in virtual environment is recommended

$ pip install -r requirements.txt

Configure

Create following directories for model, voca and dataset in root folder

$ mkdir checkpoint  
$ mkdir dataset     

The rest configuration can be updated in config.py

Data preprocessing:

Install Ghidra open-source disassembler update GHIDRA_ANALYZEHEADLESS_PATH and BINARY_PATH in preprocess/run.py

$ cd preprocess/
$ python run.py

Next for preprocessing run preprocess.py

$ python preprocess.py

Modeling BinAdapter:

adding adapter to Asmdepictor

model = BinAdapter(pretrained_model)

loading adapter and embeddings (if saved)

model.load_state_dict(torch.load(
    '/path/to/adapter/state/'), strict=False)

model.module.encoder.src_word_emb = torch.load(
    '/path/to/src_embedding/')