CI: pre-commit docformatter (#63)

* add pre-commit docformatter
* fix args
* fix deprecated

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

Author: Borda

.actions/helpers.py

@@ -148,6 +148,7 @@ def _meta_file(folder: str) -> str:
     @staticmethod
     def augment_script(fpath: str):
         """Add template header and footer to the python base script.
+
         Args:
             fpath: path to python script
         """
@@ -313,6 +314,7 @@ def parse_requirements(dir_path: str):
     @staticmethod
     def copy_notebooks(path_root: str, path_docs_ipynb: str = "docs/source/notebooks"):
         """Copy all notebooks from a folder to doc folder.
+
         Args:
             path_root: source path to the project root in this tutorials
             path_docs_ipynb: destination path to the notebooks location
@@ -362,7 +364,7 @@ def update_env_details(dir_path: str):
         req = [r.strip() for r in req]
 
         def _parse(pkg: str, keys: str = " <=>") -> str:
-            """Parsing just the package name"""
+            """Parsing just the package name."""
             if any(c in pkg for c in keys):
                 ix = min(pkg.index(c) for c in keys if c in pkg)
                 pkg = pkg[:ix]

.pre-commit-config.yaml

@@ -29,6 +29,12 @@ repos:
         args: [--py36-plus]
         name: Upgrade code
 
+  - repo: https://github.com/myint/docformatter
+    rev: v1.4
+    hooks:
+      - id: docformatter
+        args: [--in-place, --wrap-summaries=115, --wrap-descriptions=120]
+
   - repo: https://github.com/PyCQA/isort
     rev: 5.9.2
     hooks:

course_UvA-DL/autoregressive-image-modeling/Autoregressive_Image_Modeling.py

@@ -184,14 +184,14 @@ def show_imgs(imgs):
 class MaskedConvolution(nn.Module):
 
     def __init__(self, c_in, c_out, mask, **kwargs):
-        """
-        Implements a convolution with mask applied on its weights.
-        Inputs:
-            c_in - Number of input channels
-            c_out - Number of output channels
-            mask - Tensor of shape [kernel_size_H, kernel_size_W] with 0s where
+        """Implements a convolution with mask applied on its weights.
+
+        Args:
+            c_in: Number of input channels
+            c_out: Number of output channels
+            mask: Tensor of shape [kernel_size_H, kernel_size_W] with 0s where
                    the convolution should be masked, and 1s otherwise.
-            kwargs - Additional arguments for the convolution
+            kwargs: Additional arguments for the convolution
         """
         super().__init__()
         # For simplicity: calculate padding automatically
@@ -290,12 +290,12 @@ def __init__(self, c_in, c_out, kernel_size=3, mask_center=False, **kwargs):
 
 
 def show_center_recep_field(img, out):
-    """
-    Calculates the gradients of the input with respect to the output center pixel,
-    and visualizes the overall receptive field.
-    Inputs:
-        img - Input image for which we want to calculate the receptive field on.
-        out - Output features/loss which is used for backpropagation, and should be
+    """Calculates the gradients of the input with respect to the output center pixel, and visualizes the overall
+    receptive field.
+
+    Args:
+        img: Input image for which we want to calculate the receptive field on.
+        out: Output features/loss which is used for backpropagation, and should be
               the output of the network/computation graph.
     """
     # Determine gradients
@@ -476,9 +476,7 @@ def show_center_recep_field(img, out):
 class GatedMaskedConv(nn.Module):
 
     def __init__(self, c_in, **kwargs):
-        """
-        Gated Convolution block implemented the computation graph shown above.
-        """
+        """Gated Convolution block implemented the computation graph shown above."""
         super().__init__()
         self.conv_vert = VerticalStackConvolution(c_in, c_out=2 * c_in, **kwargs)
         self.conv_horiz = HorizontalStackConvolution(c_in, c_out=2 * c_in, **kwargs)
@@ -558,10 +556,10 @@ def __init__(self, c_in, c_hidden):
         self.example_input_array = train_set[0][0][None]
 
     def forward(self, x):
-        """
-        Forward image through model and return logits for each pixel.
-        Inputs:
-            x - Image tensor with integer values between 0 and 255.
+        """Forward image through model and return logits for each pixel.
+
+        Args:
+            x: Image tensor with integer values between 0 and 255.
         """
         # Scale input from 0 to 255 back to -1 to 1
         x = (x.float() / 255.0) * 2 - 1
@@ -589,11 +587,11 @@ def calc_likelihood(self, x):
 
     @torch.no_grad()
     def sample(self, img_shape, img=None):
-        """
-        Sampling function for the autoregressive model.
-        Inputs:
-            img_shape - Shape of the image to generate (B,C,H,W)
-            img (optional) - If given, this tensor will be used as
+        """Sampling function for the autoregressive model.
+
+        Args:
+            img_shape: Shape of the image to generate (B,C,H,W)
+            img (optional): If given, this tensor will be used as
                              a starting image. The pixels to fill
                              should be -1 in the input tensor.
         """

course_UvA-DL/deep-autoencoders/Deep_Autoencoders.py

@@ -133,11 +133,11 @@ def __init__(
         self, num_input_channels: int, base_channel_size: int, latent_dim: int, act_fn: object = nn.GELU
     ):
         """
-        Inputs:
-            - num_input_channels : Number of input channels of the image. For CIFAR, this parameter is 3
-            - base_channel_size : Number of channels we use in the first convolutional layers. Deeper layers might use a duplicate of it.
-            - latent_dim : Dimensionality of latent representation z
-            - act_fn : Activation function used throughout the encoder network
+        Args:
+           num_input_channels : Number of input channels of the image. For CIFAR, this parameter is 3
+           base_channel_size : Number of channels we use in the first convolutional layers. Deeper layers might use a duplicate of it.
+           latent_dim : Dimensionality of latent representation z
+           act_fn : Activation function used throughout the encoder network
         """
         super().__init__()
         c_hid = base_channel_size
@@ -195,11 +195,11 @@ def __init__(
         self, num_input_channels: int, base_channel_size: int, latent_dim: int, act_fn: object = nn.GELU
     ):
         """
-        Inputs:
-            - num_input_channels : Number of channels of the image to reconstruct. For CIFAR, this parameter is 3
-            - base_channel_size : Number of channels we use in the last convolutional layers. Early layers might use a duplicate of it.
-            - latent_dim : Dimensionality of latent representation z
-            - act_fn : Activation function used throughout the decoder network
+        Args:
+           num_input_channels : Number of channels of the image to reconstruct. For CIFAR, this parameter is 3
+           base_channel_size : Number of channels we use in the last convolutional layers. Early layers might use a duplicate of it.
+           latent_dim : Dimensionality of latent representation z
+           act_fn : Activation function used throughout the decoder network
         """
         super().__init__()
         c_hid = base_channel_size
@@ -263,17 +263,13 @@ def __init__(
         self.example_input_array = torch.zeros(2, num_input_channels, width, height)
 
     def forward(self, x):
-        """
-        The forward function takes in an image and returns the reconstructed image
-        """
+        """The forward function takes in an image and returns the reconstructed image."""
         z = self.encoder(x)
         x_hat = self.decoder(z)
         return x_hat
 
     def _get_reconstruction_loss(self, batch):
-        """
-        Given a batch of images, this function returns the reconstruction loss (MSE in our case)
-        """
+        """Given a batch of images, this function returns the reconstruction loss (MSE in our case)"""
         x, _ = batch  # We do not need the labels
         x_hat = self.forward(x)
         loss = F.mse_loss(x, x_hat, reduction="none")

course_UvA-DL/deep-energy-based-generative-models/Deep_Energy_Models.py

@@ -333,11 +333,11 @@ class Sampler:
 
     def __init__(self, model, img_shape, sample_size, max_len=8192):
         """
-        Inputs:
-            model - Neural network to use for modeling E_theta
-            img_shape - Shape of the images to model
-            sample_size - Batch size of the samples
-            max_len - Maximum number of data points to keep in the buffer
+        Args:
+            model: Neural network to use for modeling E_theta
+            img_shape: Shape of the images to model
+            sample_size: Batch size of the samples
+            max_len: Maximum number of data points to keep in the buffer
         """
         super().__init__()
         self.model = model
@@ -347,11 +347,11 @@ def __init__(self, model, img_shape, sample_size, max_len=8192):
         self.examples = [(torch.rand((1, ) + img_shape) * 2 - 1) for _ in range(self.sample_size)]
 
     def sample_new_exmps(self, steps=60, step_size=10):
-        """
-        Function for getting a new batch of "fake" images.
-        Inputs:
-            steps - Number of iterations in the MCMC algorithm
-            step_size - Learning rate nu in the algorithm above
+        """Function for getting a new batch of "fake" images.
+
+        Args:
+            steps: Number of iterations in the MCMC algorithm
+            step_size: Learning rate nu in the algorithm above
         """
         # Choose 95% of the batch from the buffer, 5% generate from scratch
         n_new = np.random.binomial(self.sample_size, 0.05)
@@ -369,14 +369,14 @@ def sample_new_exmps(self, steps=60, step_size=10):
 
     @staticmethod
     def generate_samples(model, inp_imgs, steps=60, step_size=10, return_img_per_step=False):
-        """
-        Function for sampling images for a given model.
-        Inputs:
-            model - Neural network to use for modeling E_theta
-            inp_imgs - Images to start from for sampling. If you want to generate new images, enter noise between -1 and 1.
-            steps - Number of iterations in the MCMC algorithm.
-            step_size - Learning rate nu in the algorithm above
-            return_img_per_step - If True, we return the sample at every iteration of the MCMC
+        """Function for sampling images for a given model.
+
+        Args:
+            model: Neural network to use for modeling E_theta
+            inp_imgs: Images to start from for sampling. If you want to generate new images, enter noise between -1 and 1.
+            steps: Number of iterations in the MCMC algorithm.
+            step_size: Learning rate nu in the algorithm above
+            return_img_per_step: If True, we return the sample at every iteration of the MCMC
         """
         # Before MCMC: set model parameters to "required_grad=False"
         # because we are only interested in the gradients of the input.

course_UvA-DL/graph-neural-networks/GNN_overview.py

@@ -172,9 +172,9 @@ def __init__(self, c_in, c_out):
 
     def forward(self, node_feats, adj_matrix):
         """
-        Inputs:
-            node_feats - Tensor with node features of shape [batch_size, num_nodes, c_in]
-            adj_matrix - Batch of adjacency matrices of the graph. If there is an edge from i to j,
+        Args:
+            node_feats: Tensor with node features of shape [batch_size, num_nodes, c_in]
+            adj_matrix: Batch of adjacency matrices of the graph. If there is an edge from i to j,
                          adj_matrix[b,i,j]=1 else 0. Supports directed edges by non-symmetric matrices.
                          Assumes to already have added the identity connections.
                          Shape: [batch_size, num_nodes, num_nodes]
@@ -302,13 +302,13 @@ class GATLayer(nn.Module):
 
     def __init__(self, c_in, c_out, num_heads=1, concat_heads=True, alpha=0.2):
         """
-        Inputs:
-            c_in - Dimensionality of input features
-            c_out - Dimensionality of output features
-            num_heads - Number of heads, i.e. attention mechanisms to apply in parallel. The
+        Args:
+            c_in: Dimensionality of input features
+            c_out: Dimensionality of output features
+            num_heads: Number of heads, i.e. attention mechanisms to apply in parallel. The
                         output features are equally split up over the heads if concat_heads=True.
-            concat_heads - If True, the output of the different heads is concatenated instead of averaged.
-            alpha - Negative slope of the LeakyReLU activation.
+            concat_heads: If True, the output of the different heads is concatenated instead of averaged.
+            alpha: Negative slope of the LeakyReLU activation.
         """
         super().__init__()
         self.num_heads = num_heads
@@ -328,10 +328,10 @@ def __init__(self, c_in, c_out, num_heads=1, concat_heads=True, alpha=0.2):
 
     def forward(self, node_feats, adj_matrix, print_attn_probs=False):
         """
-        Inputs:
-            node_feats - Input features of the node. Shape: [batch_size, c_in]
-            adj_matrix - Adjacency matrix including self-connections. Shape: [batch_size, num_nodes, num_nodes]
-            print_attn_probs - If True, the attention weights are printed during the forward pass
+        Args:
+            node_feats: Input features of the node. Shape: [batch_size, c_in]
+            adj_matrix: Adjacency matrix including self-connections. Shape: [batch_size, num_nodes, num_nodes]
+            print_attn_probs: If True, the attention weights are printed during the forward pass
                                (for debugging purposes)
         """
         batch_size, num_nodes = node_feats.size(0), node_feats.size(1)
@@ -507,14 +507,14 @@ def __init__(
         **kwargs,
     ):
         """
-        Inputs:
-            c_in - Dimension of input features
-            c_hidden - Dimension of hidden features
-            c_out - Dimension of the output features. Usually number of classes in classification
-            num_layers - Number of "hidden" graph layers
-            layer_name - String of the graph layer to use
-            dp_rate - Dropout rate to apply throughout the network
-            kwargs - Additional arguments for the graph layer (e.g. number of heads for GAT)
+        Args:
+            c_in: Dimension of input features
+            c_hidden: Dimension of hidden features
+            c_out: Dimension of the output features. Usually number of classes in classification
+            num_layers: Number of "hidden" graph layers
+            layer_name: String of the graph layer to use
+            dp_rate: Dropout rate to apply throughout the network
+            kwargs: Additional arguments for the graph layer (e.g. number of heads for GAT)
         """
         super().__init__()
         gnn_layer = gnn_layer_by_name[layer_name]
@@ -533,9 +533,9 @@ def __init__(
 
     def forward(self, x, edge_index):
         """
-        Inputs:
-            x - Input features per node
-            edge_index - List of vertex index pairs representing the edges in the graph (PyTorch geometric notation)
+        Args:
+            x: Input features per node
+            edge_index: List of vertex index pairs representing the edges in the graph (PyTorch geometric notation)
         """
         for layer in self.layers:
             # For graph layers, we need to add the "edge_index" tensor as additional input
@@ -560,12 +560,12 @@ class MLPModel(nn.Module):
 
     def __init__(self, c_in, c_hidden, c_out, num_layers=2, dp_rate=0.1):
         """
-        Inputs:
-            c_in - Dimension of input features
-            c_hidden - Dimension of hidden features
-            c_out - Dimension of the output features. Usually number of classes in classification
-            num_layers - Number of hidden layers
-            dp_rate - Dropout rate to apply throughout the network
+        Args:
+            c_in: Dimension of input features
+            c_hidden: Dimension of hidden features
+            c_out: Dimension of the output features. Usually number of classes in classification
+            num_layers: Number of hidden layers
+            dp_rate: Dropout rate to apply throughout the network
         """
         super().__init__()
         layers = []
@@ -578,8 +578,8 @@ def __init__(self, c_in, c_hidden, c_out, num_layers=2, dp_rate=0.1):
 
     def forward(self, x, *args, **kwargs):
         """
-        Inputs:
-            x - Input features per node
+        Args:
+            x: Input features per node
         """
         return self.layers(x)
 
@@ -858,12 +858,12 @@ class GraphGNNModel(nn.Module):
 
     def __init__(self, c_in, c_hidden, c_out, dp_rate_linear=0.5, **kwargs):
         """
-        Inputs:
-            c_in - Dimension of input features
-            c_hidden - Dimension of hidden features
-            c_out - Dimension of output features (usually number of classes)
-            dp_rate_linear - Dropout rate before the linear layer (usually much higher than inside the GNN)
-            kwargs - Additional arguments for the GNNModel object
+        Args:
+            c_in: Dimension of input features
+            c_hidden: Dimension of hidden features
+            c_out: Dimension of output features (usually number of classes)
+            dp_rate_linear: Dropout rate before the linear layer (usually much higher than inside the GNN)
+            kwargs: Additional arguments for the GNNModel object
         """
         super().__init__()
         self.GNN = GNNModel(
@@ -876,10 +876,10 @@ def __init__(self, c_in, c_hidden, c_out, dp_rate_linear=0.5, **kwargs):
 
     def forward(self, x, edge_index, batch_idx):
         """
-        Inputs:
-            x - Input features per node
-            edge_index - List of vertex index pairs representing the edges in the graph (PyTorch geometric notation)
-            batch_idx - Index of batch element for each node
+        Args:
+            x: Input features per node
+            edge_index: List of vertex index pairs representing the edges in the graph (PyTorch geometric notation)
+            batch_idx: Index of batch element for each node
         """
         x = self.GNN(x, edge_index)
         x = geom_nn.global_mean_pool(x, batch_idx)  # Average pooling