Testing functional module (#50)

* Update functional tests

* Update docs styles

* Update level and circular examples

* Add type bounds test for bundle and permute

* Add encoding tests

* Add utility tests

* Use utility functions in embeddings

Author: mikeheddes

docs/_static/css/custom.css

@@ -233,10 +233,13 @@ html.writer-html5 .rst-content table.docutils th {
 }
 
 html,
-body {
-    background: #ffffff;
+body,
+.wy-body-for-nav,
+.wy-nav-content {
+    background: #ffffff !important;
 }
 
+
 .wy-nav-content-wrap {
     background: none;
 }

torchhd/embeddings.py

@@ -151,13 +151,9 @@ def reset_parameters(self):
         self._fill_padding_idx_with_zero()
 
     def forward(self, input: torch.Tensor) -> torch.Tensor:
-        # tranform the floating point input to an index
-        # make first variable a copy of the input, then we can reuse the buffer.
-        # normalized between 0 and 1
-        normalized = (input - self.low_value) / (self.high_value - self.low_value)
-
-        indices = normalized.mul_(self.num_embeddings).floor_()
-        indices = indices.clamp_(0, self.num_embeddings - 1).long()
+        indices = functional.value_to_index(
+            input, self.low_value, self.high_value, self.num_embeddings
+        ).clamp(0, self.num_embeddings - 1)
 
         return super(Level, self).forward(indices)
 
@@ -219,14 +215,9 @@ def reset_parameters(self):
         self._fill_padding_idx_with_zero()
 
     def forward(self, input: torch.Tensor) -> torch.Tensor:
-        # tranform the floating point input to an index
-        # make first variable a copy of the input, then we can reuse the buffer.
-        # normalized between 0 and 1
-        normalized = (input - self.low_value) / (self.high_value - self.low_value)
-        normalized.remainder_(1.0)
-
-        indices = normalized.mul_(self.num_embeddings).floor_()
-        indices = indices.clamp_(0, self.num_embeddings - 1).long()
+        indices = functional.value_to_index(
+            input, self.low_value, self.high_value, self.num_embeddings
+        ).remainder(self.num_embeddings - 1)
 
         return super(Circular, self).forward(indices)