Merge pull request #2 from loucerac/develop

Develop

.gitignore

@@ -11,6 +11,9 @@ out/
 *.pickle
 *.pkl
 
+## catboost
+catboost_info
+
 # Based on:
 #  https://github.com/github/gitignore/blob/master/R.gitignore
 
@@ -269,3 +272,13 @@ Temporary Items
 .apdisk
 
 notebooks/sandbox_20190416/hypmorf_fanconi_02_from_targets_to_pathways_global_explained_relevance\.tsv
+notebooks/sandbox_gpu/*.jbl
+notebooks/sandbox_gpu/*.txt
+notebooks/sandbox_gpu/*.json
+err.txt
+out.txt
+data/
+err2.txt
+out2.txt
+*.err
+*.out

.vscode/settings.json

@@ -1,8 +1,25 @@
 {
-    "python.pythonPath": "/home/cloucera/apps/miniconda3/envs/hord/bin/python",
+    "python.pythonPath": "${workspaceFolder}/.venv/bin/python",
     "python.linting.pylintEnabled": true,
-    "python.dataScience.jupyterServerURI": "http://127.0.0.1:8890",
     "autoDocstring.docstringFormat": "numpy",
     "python.linting.enabled": true,
-    "terminal.integrated.fontSize": 13
+    "terminal.integrated.fontSize": 13,
+    "python.formatting.provider": "black",
+    "editor.rulers": [88],
+    "files.watcherExclude": {
+        "**/venvs": true,
+        "**/.venv": true,
+        "**/data": true
+    },
+    "files.exclude": {
+        "**/venvs": true,
+        "**/.venv": true,
+        "**/data": true
+    },
+    "search.exclude": {
+        "**/venvs": true,
+        "**/.venv": true,
+        "**/data": true
+      }
 }
+

README.md

@@ -1,9 +1,11 @@
-# Holistic Rare Disease paper review March 2019
+# Holistic Rare Disease
 
 ## Setup
 
+In order to compute the SHAP explanations with a CUDA-enabled GPU install the `shap` change the `CUDAHOME` environment variable. Note that the `CUDAHOME` variable has already been specified in the conda environment file, see the conda version requirement.
+
+The GPU version has been tested on GNU/Linux x64 4.15 with cuda 10.2 and conda >= 4.3 . Use the following command in order to create the python environment needed to run the experiments.
+
 ```
-mv example.env .env
+conda env create -p ./.venv -f environment.yml
 ```
-
-Edit `.env` as desired.

environment.yml

@@ -1,149 +1,31 @@
-name: hord
+name: panrd
 channels:
-  - intel
-  - defaults
   - conda-forge
+  - defaults
 dependencies:
-  - gpflow=1.3.0=py36_0
-  - python-dotenv=0.10.1=py_0
-  - absl-py=0.7.0=py36_0
-  - astor=0.7.1=py36_0
-  - atomicwrites=1.3.0=py_0
-  - attrs=18.2.0=py36h28b3542_0
-  - c-ares=1.15.0=h7b6447c_1
-  - click=7.0=py36_0
-  - gast=0.2.2=py36_0
-  - grpcio=1.14.1=py36h9ba97e2_0
-  - h5py=2.8.0=py36h989c5e5_3
-  - jupyterlab=0.35.3=py36_0
-  - jupyterlab_server=0.2.0=py36_0
-  - keras-applications=1.0.6=py36_0
-  - keras-preprocessing=1.0.5=py36_0
-  - libgcc-ng=8.2.0=hdf63c60_1
-  - libprotobuf=3.6.1=hd408876_0
-  - libstdcxx-ng=8.2.0=hdf63c60_1
-  - markdown=3.0.1=py36_0
-  - more-itertools=5.0.0=py36_0
-  - multipledispatch=0.6.0=py36_0
-  - pluggy=0.9.0=py36_0
-  - protobuf=3.6.1=py36he6710b0_0
-  - py=1.8.0=py36_0
-  - pytest=3.9.1=py36_0
-  - tensorboard=1.12.2=py36he6710b0_0
-  - termcolor=1.1.0=py36_1
-  - asn1crypto=0.24.0=py36_3
-  - backcall=0.1.0=py36_2
-  - backports=1.0=py36_9
-  - bleach=2.1.3=py36_2
-  - bzip2=1.0.6=17
-  - certifi=2018.1.18=py36_2
-  - cffi=1.11.5=py36_3
-  - chardet=3.0.4=py36_3
-  - cryptography=2.3=py36_0
-  - cycler=0.10.0=py36_7
-  - cython=0.28.5=py36h1aa3f02_0
-  - daal=2019.2=intel_187
-  - daal4py=0.2019.1=py36_1
-  - decorator=4.3.0=py36_3
-  - entrypoints=0.2.3=py36_2
-  - freetype=2.9=3
-  - get_terminal_size=1.0.0=py36_7
-  - hdf5=1.10.2=2
-  - html5lib=1.0.1=py36_4
-  - icc_rt=2019.2=intel_187
-  - idna=2.6=py36_3
-  - impi_rt=2019.2=intel_187
-  - intel-openmp=2019.2=intel_187
-  - intelpython=2019.1=0
-  - intelpython3_core=2019.1=1
-  - intelpython3_full=2019.1=1
-  - ipp=2019.2=intel_187
-  - ipykernel=4.6.1=py36_2
-  - ipyparallel=6.0.2=py36_2
-  - ipython=6.3.1=py36_3
-  - ipython_genutils=0.2.0=py36_2
-  - ipywidgets=7.0.0=py36_2
-  - jedi=0.12.0=py36_2
-  - jinja2=2.9.6=py36_2
-  - jsonschema=2.6.0=py36_2
-  - jupyter=1.0.0=py36_7
-  - jupyter_client=5.1.0=py36_5
-  - jupyter_console=5.1.0=py36_2
-  - jupyter_core=4.4.0=py36_6
-  - kiwisolver=1.0.1=py36_2
-  - libffi=3.2.1=11
-  - libpng=1.6.35=0
-  - libsodium=1.0.16=3
-  - llvmlite=0.25.0=py36_2
-  - markupsafe=1.0=py36_3
-  - matplotlib=2.2.3=py36_0
-  - mistune=0.8.3=py36_2
-  - mkl=2019.2=intel_187
-  - mkl-service=1.0.0=py36_6
-  - mkl_fft=1.0.6=py36_1
-  - mkl_random=1.0.1=py36_9
-  - mpi4py=3.0.0=py36_3
-  - mpmath=1.0.0=py36_2
-  - nbconvert=5.2.1=py36_2
-  - nbformat=4.4.0=py36_2
-  - nose=1.3.7=py36_18
-  - notebook=5.2.2=py36_1
-  - numba=0.40.0=np115py36_2
-  - numexpr=2.6.8=py36_0
-  - numpy=1.15.4=py36_2
-  - numpy-base=1.15.4=py36_2
-  - openssl=1.0.2p=0
-  - pandas=0.23.4=py36_1
-  - pandocfilters=1.4.1=py36_2
-  - parso=0.2.0=py36_2
-  - path.py=11.0.1=py36_2
-  - pexpect=4.2.1=py36_4
-  - pickleshare=0.7.4=py36_3
-  - pip=9.0.3=py36_1
-  - prompt_toolkit=1.0.15=py36_2
-  - ptyprocess=0.5.2=py36_2
-  - pycparser=2.18=py36_2
-  - pyeditline=2.0.0=py36_0
-  - pygments=2.2.0=py36_5
-  - pyopenssl=17.5.0=py36_2
-  - pyparsing=2.2.0=py36_2
-  - pysocks=1.6.7=py36_1
-  - pytables=3.4.3=np115py36_3
-  - python=3.6.5=11
-  - python-dateutil=2.6.0=py36_12
-  - pytz=2018.4=py36_3
-  - pyyaml=4.1=py36_2
-  - pyzmq=16.0.2=py36_6
-  - requests=2.19.1=py36_0
-  - scikit-learn=0.20.0=py36_13
-  - scipy=1.1.0=np115py36_6
-  - setuptools=39.0.1=py36_0
-  - simplegeneric=0.8.1=py36_7
-  - six=1.11.0=py36_3
-  - smp=0.1.4=py36_0
-  - sqlite=3.23.1=1
-  - sympy=1.3=py36_1
-  - tbb=2019.3=intel_0
-  - tbb4py=2019.3=py36_intel_0
-  - tcl=8.6.4=20
-  - tensorflow=1.12.0=py36_0
-  - terminado=0.8.1=py36_2
-  - testpath=0.3.1=py36_2
-  - tk=8.6.4=28
-  - tornado=4.5.2=py36_5
-  - traitlets=4.3.2=py36_3
-  - urllib3=1.22=py36_2
-  - wcwidth=0.1.7=py36_6
-  - webencodings=0.5.1=py36_0
-  - werkzeug=0.14.1=py36_0
-  - wheel=0.31.0=py36_3
-  - widgetsnbextension=3.2.0=py36_1
-  - xgboost=0.7=py36_0
-  - xz=5.2.3=2
-  - yaml=0.1.7=2
-  - zeromq=4.2.3=2
-  - zlib=1.2.11=5
+  - _libgcc_mutex=0.1
+  - _openmp_mutex=4.5
+  - ca-certificates=2020.12.5
+  - certifi=2020.12.5
+  - ld_impl_linux-64=2.35.1
+  - libffi=3.3
+  - libgcc-ng=9.3.0
+  - libgomp=9.3.0
+  - libstdcxx-ng=9.3.0
+  - ncurses=6.2
+  - openssl=1.1.1i
+  - pip=20.3.3
+  - python=3.8.6
+  - python_abi=3.8
+  - readline=8.0
+  - setuptools=49.6.0
+  - sqlite=3.34.0
+  - tk=8.6.10
+  - wheel=0.36.2
+  - xz=5.2.5
+  - zlib=1.2.11
+  - git
   - pip:
-    - backports.shutil-get-terminal-size==1.0.0
-    - tables==3.4.3
-
+    - -r file:requirements.txt
+variables:
+  CUDAHOME: /usr/local/cuda-10.2/