Black-System-Resource-Modelling

Source code for the paper entitled "A Trace-Driven Approach to Mobile App Functionality Classification"

Environment

Create the Environment

Step 1: conda create --name rat python=3.6.5;

Step 2: activate the environment 'rat' by running 'conda activate rat';

Step 3: install the following libraries in 'rat'.

    Numpy: conda install -c anaconda numpy=1.14.3
    Pandas: conda install -c anaconda pandas=0.23.0
    Scikit-learn: conda install -c anaconda scikit-learn=0.19.1
    Treelib: conda install -c conda-forge treelib=1.5.5
    Gensim: conda install -c anaconda gensim=3.4.0
    Tensorflow: 
        if you have a GPU equiped with CUDA:
            conda install tensorflow=1.11.0=gpu_py36h5dc63e2_0
        else:
            conda install tensorflow=1.10.0=eigen_py36h849fbd8_0
    H5Py: conda install -c anaconda h5py=2.9.0
    Matplotlib: conda install -c conda-forge matplotlib=2.2.2
    Wordcloud: conda install -c conda-forge wordcloud

Export the Environment

Step 1: activate the environment 'rat' by running 'conda activate rat';

Step 2: conda env export --file rat_20210627.yml

Step 3: find and store the environment file 'rat_20210627.yml' as shown in the project directory.

Import the Environment

Step 1: move 'rat_20210627.yml' to 'd:' of the target computer;

Step 2: conda env create -f d:\rat_20210627.yml.