Python QC workflow (Execute Order #66)

* Add initial linting workflow

* Update workflow - no self installation

* Update to make pydocstyle --numpy compliant.

* Rename workflow to python_qc, remove mypy and pylint.

* Update to only apply to PRs to main.

.github/workflows/python_qc.yaml

@@ -0,0 +1,32 @@
+name: python-code-qc
+
+on:
+  # run on every push to main
+  push:
+    branches:
+    - main
+  # run on every push (not commit) to a PR, plus open/reopen
+  pull_request:
+    types:
+    - synchronize
+    - opened
+    - reopened
+
+jobs:
+  various:
+    name: Python Code QC (Black, pydocstyle)
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v3
+    - uses: actions/setup-python@v4
+      with:
+        python-version: "3.11"
+    - run: pip install black pydocstyle
+
+    # annotate each step with `if: always` to run all regardless
+    - name: Assert that code matches Black code style
+      if: always()
+      uses: psf/black@stable
+    - name: Lint with pydocstyle
+      if: always()
+      run: pydocstyle --convention=numpy ./

examples/1_SimpleNet/simplenet.py

@@ -1,4 +1,4 @@
-"""Module defining a simple PyTorch 'Net' for coupling to Fortran. """
+"""Module defining a simple PyTorch 'Net' for coupling to Fortran."""
 import torch
 from torch import nn
 
@@ -15,7 +15,6 @@ def __init__(
         Consists of a single Linear layer with weights predefined to
         multiply the input by 2.
         """
-
         super().__init__()
         self._fwd_seq = nn.Sequential(
             nn.Linear(5, 5, bias=False),
@@ -38,7 +37,6 @@ def forward(self, batch: torch.Tensor) -> torch.Tensor:
             batch scaled by 2.
 
         """
-
         return self._fwd_seq(batch)
 
 

examples/1_SimpleNet/simplenet_infer_python.py

@@ -21,7 +21,6 @@ def deploy(saved_model: str, device: str, batch_size: int = 1) -> torch.Tensor:
     output : torch.Tensor
         result of running inference on model with example Tensor input
     """
-
     input_tensor = torch.tensor([0.0, 1.0, 2.0, 3.0, 4.0]).repeat(batch_size, 1)
 
     if device == "cpu":

examples/2_ResNet18/resnet18.py

@@ -21,7 +21,6 @@ def initialize(precision: torch.dtype) -> torch.nn.Module:
     model: torch.nn.Module
         Pretrained ResNet-18 model
     """
-
     # Set working precision
     torch.set_default_dtype(precision)
 
@@ -96,7 +95,7 @@ def run_model(model: torch.nn.Module, precision: type) -> None:
 
 
 def print_top_results(output: torch.Tensor) -> None:
-    """Prints top 5 results
+    """Print top 5 results.
 
     Parameters
     ----------

examples/n_c_and_cpp/resnet18.py

@@ -14,7 +14,6 @@ def initialize():
     -------
     model : torch.nn.Module
     """
-
     # Load a pre-trained PyTorch model
     print("Loading pre-trained ResNet-18 model...", end="")
     model = torchvision.models.resnet18(pretrained=True)
@@ -35,7 +34,6 @@ def run_model(model):
     ----------
     model : torch.nn.Module
     """
-
     print("Running ResNet-18 model for ones...", end="")
     dummy_input = torch.ones(1, 3, 224, 224)
     output = model(dummy_input)

examples/n_c_and_cpp/resnet_infer_python.py

@@ -22,7 +22,6 @@ def deploy(saved_model, device, batch_size=1):
     output : torch.Tensor
         result of running inference on model with Tensor of ones
     """
-
     image_filename = "data/dog.jpg"
     input_image = Image.open(image_filename)
     preprocess = torchvision.transforms.Compose(