Minor fixes in docs

src/tf_tabular/builder.py

@@ -71,7 +71,7 @@ def add_inputs_list(
             )
 
     def build_input_layers(self) -> tuple[List[tf.keras.layers.Layer], tf.Tensor]:
-        """Build input layer stack and return the input layers and the output layer for building the model.
+        """Builds input layer stack and return the input layers and the output layer for building the model.
         :return tuple[List[tf.keras.layers.Layer], tf.Tensor]: Tuple containing the input layers and the output layer
         """
         input_layers = []

src/tf_tabular/numeric_processor.py

@@ -21,7 +21,7 @@ def __init__(
         self.builder = builder
         self.projection_activation = projection_activation
 
-    def project(self, x: List[tf.keras.layers.Layer]):
+    def project(self, x: List[tf.Tensor]) -> List[tf.Tensor]:
         """If num_projection is not None, project the numerical features to a lower or higher dimension.
         If a builder is provided, use that to build the projection layer. Otherwise, use the default projection layer.
         """

src/tf_tabular/utils.py

@@ -1,4 +1,6 @@
 from itertools import combinations
+from typing import List
+
 import numpy as np
 import pandas as pd
 import tensorflow as tf
@@ -17,7 +19,7 @@
 
 
 def _input_layer(name: str, is_multi_hot: bool = False, is_string: bool = False):
-    """Build input layer for a column"""
+    """Builds input layer for a column"""
     shape: tuple[None] | tuple[int]
     if is_multi_hot:
         shape = (None,)
@@ -47,7 +49,18 @@ def get_combiner(combiner: str, is_list: bool):
         raise NotImplementedError(f"Unknown combiner: {combiner}")
 
 
-def build_projection_layer(cont_layers, num_projection, l2_reg, activation="relu", cross_features=True):
+def build_projection_layer(cont_layers: List[tf.Tensor], num_projection: int, l2_reg: float,
+                           activation: str = "relu", cross_features: bool = True):
+    """Builds a projection layer for continuous features. If cross_features is True, it will also include the
+    multiplication of all pairs of continuous features.
+
+    :param List[tf.Tensor] cont_layers: List of continuous layers
+    :param int num_projection: size of projection layer output neurons
+    :param float l2_reg: regularization parameter for L2
+    :param str activation: activation to use in projection layer, defaults to "relu"
+    :param bool cross_features: Whether to build cross features or not, defaults to True
+    :return Tensor: output of the projection layer
+    """
     if cross_features:
         cont_layers = list(cont_layers)
         pairs = list(combinations(cont_layers, 2))
@@ -123,7 +136,7 @@ def get_embedding_layer(
 
 
 def build_continuous_input(name, mean: float | None = None, variance: float | None = None, sample=None):
-    """Build th e input layer stack for continuous features
+    """Builds the input layer stack for continuous features
 
     :param str name: Layer name
     :param float mean: mean of the feature values, defaults to None
@@ -142,7 +155,7 @@ def build_continuous_input(name, mean: float | None = None, variance: float | No
 
 
 def build_categorical_input(name, embedding_dim, vocab, is_multi_hot, embedding_df=None):
-    """Build input for categorical columns.
+    """Builds input for categorical columns.
     This function supports many cases because of the different trials we have done for different columns
 
     :param str name: Layer name
@@ -174,7 +187,7 @@ def build_categorical_input(name, embedding_dim, vocab, is_multi_hot, embedding_
 
 
 def get_vocab(series: pd.Series, max_size: int | None = None):
-    """Get the vocabulary of a series"""
+    """Gets the vocabulary (unique items) of a series"""
     if isinstance(series.iloc[0], list) or isinstance(series.iloc[0], np.ndarray):
         series = series.explode()
     series = series.dropna()