Merge branch 'develop' of https://github.com/CrayLabs/SmartSim into s…

…lurm-het-jobs

.github/workflows/run_tests.yml

@@ -108,7 +108,7 @@ jobs:
       - name: Install SmartSim (with ML backends)
         run: |
           python -m pip install git+https://github.com/CrayLabs/SmartRedis.git@develop#egg=smartredis
-          python -m pip install .[dev,ml]
+          python -m pip install .[dev,ml-cpu]
 
       - name: Install ML Runtimes with Smart (with pt, tf, and onnx support)
         if: (matrix.py_v != '3.10')

.pylintrc

@@ -341,3 +341,10 @@ max-try-statements=7
 
 # Maximum level of inheritance, bumping up to account for db mixins
 max-parents=25
+
+
+[TYPECHECK]
+
+# Pylint is kinda rough at determining what is and is not an attr in C extensions
+# leave this work to mypy, which tends to be more reliable
+ignored-modules=torch

doc/changelog.rst

@@ -19,7 +19,10 @@ To be released at some future point in time
 
 Description
 
+- Mitigated additional suppressed pylint errors
 - Fix colocated db preparation bug when using `JsrunSettings`
+- Enabled additional mypy checks
+- Additional typehints
 - Log ignored error messages from `sacct`
 - Fix malformed logging format strings
 - Fix bug when user specify CPU and devices greater than 1
@@ -28,14 +31,20 @@ Description
 
 Detailed Notes
 
+- Created public proprties where appropriate to mitigate `protected-access` errors (PR341_)
 - Fix a failure to execute `_prep_colocated_db` due to incorrect named attr check (PR339_)
+- Enabled and mitigated mypy `disallow_any_generics` and `warn_return_any` (PR338_)
+- Add typehints to `smartsim._core.launcher.step.*` (PR334_)
 - Log errors reported from slurm WLM when attempts to retrieve status fail (PR331_, PR332_)
 - Fix incorrectly formatted positional arguments in log format strings (PR330_)
 - Add error if user calls get_allocation with reserved keywords in slurm get_allocation (PR325_)
 - Add error when user requests CPU with devices greater than 1 within add_ml_model and add_script (PR324_)
 - Update pylint dependency, update .pylintrc, mitigate non-breaking issues, suppress api breaks (PR311_)
 
+.. _PR341: https://github.com/CrayLabs/SmartSim/pull/341
 .. _PR339: https://github.com/CrayLabs/SmartSim/pull/339
+.. _PR338: https://github.com/CrayLabs/SmartSim/pull/338
+.. _PR334: https://github.com/CrayLabs/SmartSim/pull/334
 .. _PR332: https://github.com/CrayLabs/SmartSim/pull/332
 .. _PR331: https://github.com/CrayLabs/SmartSim/pull/331
 .. _PR330: https://github.com/CrayLabs/SmartSim/pull/330

doc/installation_instructions/basic.rst

@@ -18,7 +18,7 @@ Basic
 
 The base prerequisites to install SmartSim and SmartRedis are:
 
-  - Python 3.7-3.9
+  - Python 3.8-3.10
   - Pip
   - Cmake 3.13.x (or later)
   - C compiler
@@ -41,7 +41,7 @@ GPU Support
 The machine-learning backends have additional requirements in order to
 use GPUs for inference
 
-  - `CUDA Toolkit 11 (11.4 recommended) <https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html>`_
+  - `CUDA Toolkit 11 (tested with 11.8) <https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html>`_
   - `cuDNN 8 (tested with 8.2.1 and 8.4.0) <https://docs.nvidia.com/deeplearning/cudnn/install-guide/index.html#download>`_
   - OS: Linux
   - GPU: Nvidia
@@ -66,11 +66,11 @@ Supported Versions
    * - MacOS
      - x86_64
      - Not supported
-     - 3.7 - 3.9
+     - 3.8 - 3.10
    * - Linux
      - x86_64
      - Nvidia
-     - 3.7 - 3.9
+     - 3.8 - 3.10
 
 
 .. note::
@@ -81,7 +81,7 @@ Supported Versions
 
 Native support for various machine learning libraries and their
 versions is dictated by our dependency on RedisAI_ 1.2.7. Users
-can also select RedisAI 1.2.3 or 1.2.5 (though that also limits
+can also select RedisAI 1.2.5 (though that also limits
 the version of the ML libraries).
 
 +------------------+----------+-------------+---------------+
@@ -91,8 +91,6 @@ the version of the ML libraries).
 +------------------+----------+-------------+---------------+
 | 1.2.5            | 1.9.0    | 2.6.0       | 1.9.0         |
 +------------------+----------+-------------+---------------+
-| 1.2.3            | 1.7.0    | 2.5.2       | 1.9.0         |
-+------------------+----------+-------------+---------------+
 
 TensorFlow_ 2.0 and Keras_ are supported through `graph freezing`_.
 
@@ -116,10 +114,10 @@ MacOS-only
 ==========
 
 We recommend users and contributors install brew_ for managing installed
-packages.  For contributors, the following brew packages can be helpful:
+packages. For contributors, the following brew packages can be helpful:
 
 - openmpi_ for building and running parallel SmartRedis examples
-- doxygen_ for building the documention
+- doxygen_ for building the documentation
 - cmake_ for building SmartSim and SmartRedis from source
 
 .. _brew: https://brew.sh/
@@ -146,7 +144,8 @@ There are two stages for the installation of SmartSim.
 Step 1: Install Python Package
 ==============================
 
-We first recommend creating a new virtual environment:
+We first recommend creating a new
+`virtual environment <https://docs.python.org/3/library/venv.html>`_:
 
 .. code-block:: bash
 
@@ -159,16 +158,17 @@ and install SmartSim from PyPI with the following command:
 
     pip install smartsim
 
-If you would like SmartSim to also install Machine Learning libraries that
-can be used outside SmartSim to build SmartSim-compatible models, you
-can request their installation through the ``ml`` flag as follows:
+If you would like SmartSim to also install python machine learning libraries
+that can be used outside SmartSim to build SmartSim-compatible models, you
+can request their installation through the ``ml-*`` optional dependencies,
+as follows:
 
 .. code-block:: bash
 
-    # For bash
-    pip install smartsim[ml]
-    # or if using ZSH
-    pip install smartsim\[ml\]
+    # For CPU based models
+    pip install smartsim[ml-cpu]
+    # For CPU and CUDA based models
+    pip install smartsim[ml-cuda]
 
 At this point, SmartSim is installed and can be used for more basic features.
 If you want to use the machine learning features of SmartSim, you will need
@@ -187,7 +187,7 @@ To see all the installation options:
 
 .. code-block:: bash
 
-    smart
+    smart --help
 
 CPU Install
 -----------
@@ -198,7 +198,7 @@ To install the default ML backends for CPU, run
 
     # run one of the following
     smart build --device cpu          # install PT and TF for cpu
-    smart build --device cpu --onnx   # install all backends (PT, TF, ONNX) on gpu
+    smart build --device cpu --onnx   # install all backends (PT, TF, ONNX) on cpu
 
 By default, ``smart`` will install PyTorch and TensorFlow backends
 for use in SmartSim.
@@ -247,9 +247,9 @@ SmartSim does.
    * - Platform
      - Python Versions
    * - MacOS
-     - 3.7 - 3.9
+     - 3.7 - 3.10
    * - Linux
-     - 3.7 - 3.9
+     - 3.7 - 3.10
 
 The Python client for SmartRedis is installed through ``pip`` as follows:
 
@@ -265,9 +265,9 @@ From Source
 ===========
 
 This section will be geared towards contributors who want to install SmartSim
-and SmartRedis from source. If you are installing from source for other reasons,
-follow the steps below but use the distribution provided hosted on GitHub or
-PyPI.
+and SmartRedis from source for development purposes. If you are installing
+from source for other reasons, follow the steps below but use the source
+distributions provided on GitHub or PyPI.
 
 .. _from-source:
 
@@ -281,14 +281,15 @@ First, clone SmartSim.
   git clone https://github.com/CrayLabs/SmartSim smartsim
 
 And then install SmartSim with pip in *editable* mode. This way, SmartSim is
-installed in your virtual environment and available in PYTHONPATH, but the
+installed in your virtual environment and available on `sys.path`, but the
 source remains at the site of the clone instead of in site-packages.
 
 .. code-block:: bash
 
   cd smartsim
-  pip install -e .[dev,ml]   # for bash users
-  pip install -e .\[dev,ml\] # for zsh users
+  pip install -e .[dev,ml-cpu]   # for CPU only
+  # OR
+  pip install -e .[dev,ml-cuda]  # for CUDA support
 
 Use the now installed ``smart`` cli to install the machine learning runtimes.
 
@@ -306,4 +307,3 @@ Build the SmartRedis library
 .. include:: ../../smartredis/doc/install/lib.rst
 
 
-

doc/installation_instructions/platform/cray.rst

@@ -8,7 +8,7 @@ if these instructions do not work).
 .. code-block:: bash
 
     module use -a /lus/scratch/smartsim/local/modulefiles
-    module load cudatoolkit/11.4 cudnn git-lfs
+    module load cudatoolkit/11.8 cudnn git-lfs
 
     module unload PrgEnv-cray PrgEnv-intel PrgEnv-gcc
     module load PrgEnv-gnu

doc/installation_instructions/site-install.rst

@@ -10,6 +10,6 @@ from source with the following steps replacing ``COMPILER_VERSION`` and
 .. code:: bash
 
     module use -a /lus/scratch/smartsim/local/modulefiles
-    module load cudatoolkit/11.4 cudnn smartsim-deps/COMPILER_VERSION/SMARTSIM_VERSION
-    pip install smartsim[ml]
+    module load cudatoolkit/11.8 cudnn smartsim-deps/COMPILER_VERSION/SMARTSIM_VERSION
+    pip install smartsim[ml-cuda]
     smart build --only_python_packages --device gpu [--onnx]

pyproject.toml

@@ -101,7 +101,6 @@ module = [
   "redis.cluster",
   "keras",
   "torch",
-  "smartsim._core.launcher.step.*", # hints incomplete
   "smartsim.ml.torch.*",            # must solve/ignore inheritance issues
 ]
 ignore_missing_imports = true

setup.py

@@ -191,7 +191,7 @@ def has_ext_modules(_placeholder):
         "types-setuptools",
     ],
     # see smartsim/_core/_install/buildenv.py for more details
-    "ml": versions.ml_extras_required(),
+    **versions.ml_extras_required()
 }
 
 

smartsim/_core/_cli/__main__.py

@@ -28,6 +28,7 @@
 
 from smartsim._core._cli.cli import default_cli
 
+
 def main() -> int:
     smart_cli = default_cli()
     return smart_cli.execute(sys.argv)